Vinylogous sulfonamides in the total synthesis of indolizidine alkaloids from amphibians and ants

Abstract

This thesis describes the application of vinylogous sulfonamides in a generalised synthetic protocol for the synthesis of indolizidine alkaloids, viz. monomorine I, 5-epi-monomorine I and the key precursor to indolizidine 209D. Chapter one puts the work into perspective with a review of the different classes of amphibian alkaloids, with specific emphasis on previous syntheses of indolizidine 209D and monomorine I. This is followed by a brief overview of previous synthetic strategies employed for alkaloid synthesis in the Wits laboratories and an introduction to vinylogous sulfonamides. Chapter 2 concludes with our aims and proposed strategies for the project. The attempted total synthesis of (−)-indolizidine 209D is described in Chapter 3. The initial three steps to prepare t-butyl (3R)-3-{benzyl[(1R)-1- phenylethyl]amino}nonanoate (274) proceeded well, but the fourth step, deprotecting the nitrogen, gave inconsistent results and hindered the completion of the synthesis. The free amine that we succeeded in isolating, tbutyl (3R)-3-aminononanoate (275), reacted with chlorobutyryl chloride to give us lactam, t-butyl (3R)-3-(2-oxo-1-pyrrolidinyl)nonanoate (277) in addition to the unusual by-product N-(cyclopropanecarbonyl)cyclopropanecarboxamide (327). From the lactam (277) we successfully prepared the key intermediate, vinylogous sulfonamide t-butyl (3R)-3-{2-[(E)-(p-toluenesulfonyl)methylene-1- pyrrolidinyl} nonanoate (280). The vinylogous sulfonamide effectively facilitated a high-yielding cyclisation reaction to produce the bicyclic hexahydroindolizine (282). Unfortunately the failing debenzylation reaction prevented the completion of the synthesis as no more material was available. The total syntheses of (±)-monomorine I and (±)-5-epi-monomorine I are described in Chapter 4. Notable intermediates include the enamide, ethyl 3- [(2E)-2-butylidene-5-oxopyrrolidinyl]butanoate (293), which we prepared from a condensation reaction between the ketoester (292) and the racemic amine (291). The diastereoselective reduction of the enamide (293) was optimised to give ethyl 3-(2-butyl-5-oxo-1-pyrrolidinyl)butanoate (294) as a 1:5 mixture of isomers. After thionation, the two isomers were separable, (295A) was the intermediate for (±)-monomorine I and (295B) the intermediate for (±)-5-epimonomorine I. Following the formation of the vinylogous sulfonamide, the key cyclisation step proceeded well for both the diastereomers to give the hexahydroindolizines (298A) and (298B). We obtained crystal structures of both hexahydroindolizines and were able to confirm the relative stereochemistry of the isomers. Defunctionalisation of the vinylogous sulfonamide included a stereoselective platinum-catalysed reduction of the alkene, followed by desulfonylation. Conditions were optimized and the synthesis was completed to give (±)-monomorine I in an overall yield of 3% and (±)-5-epi-monomorine I in an overall yield of 7%. Approaches towards the enantioselective synthesis were explored but, unfortunately, we experienced difficulties with the debenzylation reaction required to produce the chiral amine (291). In the process of trying to circumvent this problem a side route involving monobenzylated analogues was investigated. While the side route produced some interesting products, we were unable to direct the synthetic path back towards enantiopure monomorine I. The feasibility of extending this methodology to more complex alkaloids was briefly investigated. Initial experimentation involving allylated analogues of the ketoester (292) was investigated and was found to be incompatible with our reaction conditions.