Chasing my Mavericks: Stereoselective Allylations, Cascade Reactions and Total Synthesis of Natural Products

Over billions of years, Nature has developed and optimized enantiomerically pure molecules capable of being assembled and precisely perform a variety of specific tasks.Due to the broad applicability and the long lasting interest, asymmetric allylic alkylations are one of the most vibrant and investigated fields in organic synthesis. Chapter 1 presents an overview of asymmetric allylations catalyzed by palladium. To become more familiar with palladium catalyzed allylic alkylation, we initially explored an enantioselective cyclization of Ugi adducts (Chapter 2). To access biologically relevant diketopiperazines (DKPs), we specifically design the precursors bearing an allylic carbonate, which could be later employed as the substrate for the Tsuji-Trost allylation. After a thorough optimization, we obtained a variety of DKPs in high enantioselectivity and generally high yields. The mild reaction conditions could tolerate a wide array of functional groups, and could be further decorated by several post transformation reactions. By simply replacing ketones with heterocyclic aldehydes in the synthesis of the precursors, we obtained a different class of Ugi adducts, bearing three different nucleophilic positions (N-amide, N-pyridine, C-benzylic position). To our surprise, formation of β-lactams outcompeted all other pathways. Moreover, we soon discovered that monodenate ligands promoted the formation of the trans-diastereoisomer, while the cis one was the main product with bidentate ligands. Systematic investigation revealed that intramolecular coordination of Pd by pyridine is the cause of the high diastereoselectivity observed with monodentate ligands, while the preference of the bidentate ligands is possibly caused by atypical hydrogen bonding with the ligand. Further investigation revealed that many other heterocycles (such as oxazoles, imidazoles and isoquinolines) could coordinate to the π-allyl complex. Moving to intermolecular transformations (Chapter 4), we considered the well- known donor-acceptor cyclopropanes (DACs), traditionally employed in cycloaddition reactions. However, due to the interconnected nature of the mechanism, it is extremely rare to achieve two separate, independent reactivities of the donor and acceptor moieties. Therefore, we employed vinylcyclopropanes (VCPs) bearing a phosphonate group, capable of undergoing Horner-Wadsworth-Emmons (HWE) olefination. Using salicylaldehydes, we developed of a novel chiral phosphoramidate ligand also allowed to generate a wide range of enantioenriched benzoxepins in high yields and with good enantioselectivities. Once we established the reactivity of VCPs, we decided to further investigate the scope of the reaction, focusing our interest on aromatic aldehydes bearing different nucleophilic moieties (Chapter 5). We soon realized that also 2-aminobenzaldehydes, (benzo)imidazolecarboxaldehydes, (aza)indolecarboxaldehydes, and 6-formyl-2-pyridone were suitable substrates, leading to several fused heterocyclic scaffolds. Unfortunately, our attempts to develop an enantioselective transformation using chiral ligands were unsuccessful. Moving to similarly substituted arylcyclopropanes, using scandium(III) triflate, we were able to convert a broad range of indole-2-carboxaldehydes to the corresponding dihydrocarbazoles in THF at 50 °C. In our attempts to maximize conversion, we found that performing the reaction in dioxane at 100 °C instead afforded the corresponding carbazoles as sole product. While exploring the interesting reactivity of indoles and their conversion to dihydro-β-carbolines, we serendipitously observed the transformation of amides to carbazoles under typical Bischler-Napieralski cyclization (Chapter 7). We performed a thorough investigation of the reaction mechanism, discovering a complex cascade process. Once demonstrated the generality of the cascade, by generation of a library of variously substituted carbazoles, we focused our attention on a intermediate, core of numerous natural products. In our attempt to interrupt the cascade process at this step, we discovered multiple divergent pathways, each leading to carbazoles with different substituent patterns. Finally, we realized that in order to isolate the desired intermediate, it was crucial to install two ester moieties on the amide precursor (Chapter 8). Subsequent transformations allowed us to complete the total synthesis of the alkaloid akuammicine in only six steps.