Shair’s Synthesis of (+)-Cephalostatin 1 from Hecogenin Acetate and trans-Androsterone

The architectural complexity of the bis-steroidal pyrazine cephalostatin 1, along with its potent antiproliferative activity and unique biological mechanism of action, have inspired organic chemists since its isolation by Petit in the early 1990s. We have noted previously that three laboratories have completed chemical syntheses of 1. In this post, we will examine the enantioselective synthesis of cephalostatin 1 by the Shair group at Harvard.  Shair and co-workers dissected 1 into two hemispheric fragments: the eastern substructure 2 and a western half (10) derived from hecogenin acetate. The fragment 2 was prepared from trans-androsterone, a readily available 17-keto steroid. The synthetic challenges dictated by the use of this starting material are (1) the remote oxidation of the C12 position to install the 12-hydroxy functionality within 1 and (2) installation and elaboration of a stereochemically complex spiroketal system (rings E & F) wherein the spiro-fused ring junction position (C22) is in a relatively sensitive kinetic configuration. The C22-epimeric diastereomer (of 2) is stabilized by two anomeric effects as compared with the requisite structure 2 (three-dimensional structure shown above on bottom right), which contains a monoanomeric spiroketal linkage.

Remote oxidation of the C12 position is accomplished by implementation of the copper-mediated methodology of Schonecker. We have previously discussed the mechanism of this stereoselective oxidation in the context of the synthesis of cyclopamine by Giannis and co-workers. Next, a Sonogashira coupling reaction between an optically active alkyne (which contains seven of the eight carbons of the E/F spiroketal) and a C17 enol triflate generates the advanced intermediate 3, shown above. Meinwald and Liu pioneered this type of Pd-catalyzed D-ring functionalization in 1996 and an advancement was later published by Jones and co-workers in 2001. Sharpless dihydroxylation of 3 led to a cis-diol intermediate (not shown) with a-hydroxy functionality at C17. A unique oxidation of this diol with benzeneselinic anhydride then generated an a-hydroxy cyclopentenone that was reduced with triacetoxyborohydride to produce 4. The stereochemical outcome of the latter transformation is due to directing group participation of the C17 hydroxyl group, resulting in a trans-diol (4) with a C16 hydroxyl in the b-configuration. Subsequent exposure of 4 to a gold(I)-cataytic system promoted a 5-endo-dig cyclization to forge the dihydrofuran 5 with outstanding efficiency. Next, the C17 hydroxyl group is used again, in this instance to direct a Simmons-Smith cyclopropanation reaction that stereoselectively introduces the final carbon atom of the eventual spiroketal E/F system. Desilylation then precedes the critical oxidative spiroketalization reaction using NBS to deliver predominantly the desired (kinetically favored) C22-(S) isomer 7 under pH-neutral, non-equilibrating reaction conditions. Debromination under free-radical conditions gave the advanced spiroketal 8 that was finally converted to 2 by a three-step sequence that included silylation of the hindered C17 hydroxyl and adjustment of the oxidation state of C3.

Shair’s synthesis of the western half of cephalostatin 1 from hecogenin acetate will not be discussed here in great detail except to mention that the methods used to selectively oxidize the angular C18 methyl group are indeed very interesting from a mechanistic perspective. Installation of the D(14,15) olefin was accomplished concurrently in the course the C18 oxidation reaction sequence. The reorganization of the hecogenin spiroketal system involved implementation of a well-established Marker degradation procedure followed by elongation and elaboration of a functionalized pregnane (progesterone-type) system that eventually led to 10. Overall, Shair’s preparation of 10 from hecogenin acetate is conceptually interesting but somewhat linear and lengthy as compared to that of the Tian laboratory, which has reportedly synthesized multigram quantities of a related cephalostatin western domain intermediate by a relatively concise synthetic protocol. Similar to Tian’s synthesis of 1, Shair and co-workers, in their endgame, apply the Fuchs unsymmetrical pyrazine synthesis to condense the advanced steroidal precursors 9 and 10. We recently discussed the detailed mechanism of this transformation here. Finally, in a single global deprotection step, fluoride-mediated desilylation at four positions is accompanied by hydrolysis of the C12 acetate to furnish synthetic cephalostatin 1. Completed in 2009, this landmark effort constitutes the second of only three chemical syntheses of 1 that have been disclosed to date. As a reminder, the Fuchs group at Purdue University reported the first synthesis of cephalostatin 1 in 1998 and, as noted above, the most recent chemical preparation of the complex bis-steroidal pyrazine was reported by the laboratory of Weisheng Tian.