Synthesis of (–)-melazolide B, a degraded limonoid, from a natural terpene precursor

Degraded limonoids are a subclass of limonoid natural products that derive from ring-intact or ring-rearranged limonoids. Establishment of robust synthetic routes to access them could provide valuable materials to identify the simplest active pharmacophore responsible for the observed biological activities of the parent molecules. This communication delineates the development of a divergent strategy to furnish melazolide B and several other related congeners from a common keto-lactone intermediate, which was rapidly assembled from α-ionone. A chemoselective carbonyl α,β-dehydrogenation and a Wharton reduction were key strategic steps in this synthetic pathway.


Introduction
Limonoids are a large family of terpenoid natural products with more than a thousand members isolated to date [1][2][3].These secondary metabolites display a broad array of biological activities, ranging from anticancer, anti-inflammation, antifeedant, and neurological activities [1][2][3].Due to their diverse and intricate structures as well as ☆ Professor Timothy Newhouse, the corresponding author on this paper is a member of the advisory board but this author had no involvement in the peer review process used to assess this work submitted to Tetrahedron Chem.This paper was assessed and the corresponding peer review managed by Dr Chaosheng Luo, a scientific editor working on Tetrahedron Chem.
Our established approach provided a synthetic pathway to access compounds related to the DE-ring fragments, such as azedaralide (3) and pyroangolensolide (4) [12].Herein we disclose the development of synthetic routes to access compounds related to the AB-ring fragments, including (-)-melazolide B (5) and actinidiolide (6).

Results and discussion
Our initial synthetic strategy focused on a bidirectional search between the known degraded limonoids, such as 5 and 6, and our previously reported intermediate 11.The benefit of 11 as a starting material goal is that it already contains the key ring systems, quaternary center, and two stereocenters common to these degraded limonoids [19].A cyclohexanone would need to be converted to an allylic alcohol wherein the hydroxyl group has formally undergone a reductive transposition.As an added benefit, the route to 11 was robust, and involved conversion of α-ionone (8) by a three-step sequence involving a kinetic resolution via Jacobsen epoxidation [20], 1,4-hydrosilylation, and oxidative cleavage, as shown in Scheme 3A [12].Treatment of ketone 11 with KHMDS and N-phenylbis(trifluoromethanesulfonimide) resulted in the formation of an intermediate vinyl triflate (12) in 84% yield, which was then reduced to alkene 13 in 87% yield (Scheme 3A).Other reductants employed in this Pd-catalyzed reduction, including Et 3 SiH and Bu 3 SnH, were less effective in this context.Conversion of alkene 13 to an intermediate enone was achieved by allylic oxidation utilizing Mn(OAc) 3 and t-BuOOH [21].Employing alternative allylic oxidation conditions to furnish 5 or 15 directly, such as SeO 2 and Cr-based oxidants, were unsuccessful.A diastereoselective Luche reduction of the enone intermediate (14) resulted exclusively in the formation of C3-epi-melazolide B (15).
Although 13 was not a viable intermediate to melazolide B (5), considering our laboratory's lactone α,β-dehydrogenation [22], we reasoned that subjection of 13 to lactone α,β-dehydrogenation conditions could give rise to (-)-actinidiolide ( 6), an ionone-related compound that was proposed to be produced from kiwiionoside in nature [23].Indeed, dehydrogenation of the lactone functionality in 13 with our laboratory's allyl Pd-catalyzed dehydrogenation conditions revealed that the conditions originally developed for ketone dehydrogenation were most effective (the Zn(TMP) 2 and diethyl allyl phosphate system) to produce actinidiolide (6), as shown in scheme 3B [22].Employing the conditions previously developed by our laboratory for ester [24] or amide [25] dehydrogenation resulted in lower conversion and diminished yield (23% and 47% yield respectively).These results suggest that the Zn(TMP) 2 system may be more general for the dehydrogenation of other basic functionalities.
In order to obtain melazolide B (5), we undertook an alternative route through enone 17 (Scheme 4).Several ketone dehydrogenation conditions of 11 was first examined.A twostep sequence involving TMS enol ether formation and dehydrogenation was first attempted.Treatment of ketone 11 with KHMDS and TMSCl resulted in 16 in 36% yield (Scheme 4A).Although the original Saegusa-Ito oxidation conditions only led to full decomposition of the sily enol ether starting material ( 16) [26], subjection of 16 to Tsuji's modified conditions smoothly delivered the enone product (17) in 91% yield (Scheme 4A) [27].
With enone 17 in hand, we were ready to test the proposed synthesis of melazolide B (5) through Wharton reduction [31].Treatment of enone 17 with urea•H 2 O 2 effected a nucleophilic epoxidation to furnish epoxide 19 in 67% yield as a 1:1 mixture of diastereomers (Scheme 5) [12].Reduction of the mixture of diastereomers of the α,β-epoxy ketone (19 with Wharton's hydrazine protocol resulted in the formation of (-)-melazolide B (5), via the intermediacy of hydrazone 20 (Scheme 5) [31].Interestingly, during the reduction of the α,β-epoxy ketone (19) with hydrazine only the desired diastereomer of the allylic alcohol 5 was observed, whereas the undesired diastereomer was degraded.It is unclear at this juncture what those decomposition pathway or pathways are, however the presence of an electrophilic lactone was possibly a liability.This six-step sequence marks the first reported synthesis of (-)-melazolide B.

Conclusion
In summary, we have documented efficient synthesis of C3-epi-melazolide B, melazolide B and actinidiolide.Tapping into the natural terpene precursors by utilizing α-ionone as a starting material accelerated the assembly of the core bicyclic structures and led to the first reported synthesis of (-)-melazolide B. Among the reported synthesis of (-)-actinidiolide (6) Scheme 1.
Selected limonoids and structurally related compounds.