Investigating Biogenetic Hypotheses of the Securinega Alkaloids: Enantioselective Total Syntheses of Secu’amamine E/ent-Virosine A and Bubbialine

Abstract

The synthesis of the Securinega alkaloid secu'amamine E (ent-virosine A) has been accomplished for the first time in 12 steps and 8.5% overall yield. In addition, bubbialine has been prepared and characterized. These two alkaloids and bubbialidine, all featuring an azabicyclo[2.2.2]octane core, were rearranged to their azabicyclo[3.2.1]octane congeners, a framework found in many Securinega alkaloids. These experiments suggest that azabicyclo[2.2.2]octane derivatives could serve as intermediates in the biosynthesis of the rearranged azabicyclo[3.2.1]octane products.

Bio‐inspired Total Synthesis of Twelve Securinega Alkaloids: Structural Reassignments of (+)‐Virosine B and (−)‐Episecurinol A

“…[19] It is noteworthy that the bicyclic butenolide 19 does occur naturallyi nt he form of the two diastereomeric (À)-menisdaurilide (19 a)a nd (+ +)-aquilegiolide (19 b) (Scheme 1). [20] This approachc ould thus correspond to the true biosynthetic pathway toward these alkaloids.However, the diastereoselectivity of the Mannich reaction used by BusquØ,d e March and Gademann [18e,19] had the potential to afford 4o ut of 8p ossible diastereomers, and all with an exo-orientation of the methine CÀHb ond a to the nitrogen atom ( Figure 1). Thereby,t his approach might not account for the biosynthesis of all possible (neo)(nor)securinane diastereomers, almost all of which having been identified from different plant extracts.…”

“…A cyclizing reductive amination through 22 could next be implemented without strict facioselective hydride attack to elaborate the pyrrolidine or piperidine motif of all possible diastereomers of the key intermediates 17 . Subsequent intramolecular aza‐Michael cyclisations would then furnish both exo ‐ and endo ‐neo(nor)securinanes, which could next be rearranged into their (nor)securinane variants by activation of their C‐ring secondary alcohol (Scheme ), as previously reported for the syntheses of norsecurinine ( 1 ), allonorsecurinine ( 3 ) and allosecurinine ( 4 ), and (+)‐viroallosecurinine ( 4 b ) . The key aldol/dehydrogenation/reductive amination sequence we propose, whether or not it can be echoed in the biogenesis of these alkaloids, clearly offers a higher degree of structural divergence relative to that of previous biosynthesis proposals.…”

“…The first objective of this collective synthesis plan was the preparation of an intermediate that could act as a menisdaurilide ( 19 a ) surrogate. Different protected versions of this putative biosynthetic intermediate have previously been prepared by the groups of Gademann and de March in yields ranging from about 2 to 15 % over 7 to 11 steps . Our own synthesis started from 4‐hydroxyphenylacetic acid 18 , a plausible tyrosine‐derived biosynthetic precursor, which was subjected to a phenyliodine(III) diacetate‐mediated dearomatisation in MeOH to furnish the para ‐quinol methyl ether 23 (Scheme ).…”

Nonclassical ammonium ions as intermediates in cinchona alkaloid rearrangements?

“…Ions with related structures also have been proposed as intermediates in the rearrangements of Securinega alkaloids (Scheme ) . The results of our computations also indicate that these ions are classical ammonium ions, in this case less distorted than those discussed above (Figure ).…”

“…The key difference between these two proposals is illustrated in Scheme 2. Consider a model system with a carbocation adjacent to a C-N σ-bond (7). If a threemembered ring were formed via bridging of the C-N σ-bond adjacent to the carbocation center, a nonclassical ion containing a cyclically delocalized three-center twoelectron bonding array (9) would result; note that the nitrogen atom in such a structure would still bear a lone pair.…”