In our initial studies, we achieved the first synthesis of a complex pimarane natural product. The key cyclization precursor was synthesized via a highly modular approach in only six steps. The developed strategy is based on a powerful polyene cyclization that establishes five stereocenters, forms four carbon-carbon bonds, and generates four six-membered rings. The final step involves a unique transannular endo termination. This transformation was unprecedented in the chemical literature. We later established a synthetic platform for the construction of complex ent-pimarane natural products that were previously inaccessible via chemical synthesis. The developed synthesis benefits from robust transformations and allows for selective late-stage diversification. Highlights include an HFIP-mediated bicyclization to establish the trans-decalin stereochemistry and an arene hydrogenation/oxidation/alkylation sequence to install the crucial C13 quaternary stereocenter. The key intermediate was prepared in 10 steps and served as a late branching point to selectively modify the A- or C-ring of the tricyclic core. In total, eight natural products were accessible in 11-16 steps from commercially available starting materials. This work represents the first divergent total synthesis of ent-pimarane natural products. The developed synthesis allowed us to prepare sufficient amounts for biological screening, which is performed in collaboration with the Koeberle group (MPI Innsbruck). To access (neo)triperifordin, we investigated a powerful cationic polyene cyclization to reach the tricyclic carbon skeleton. In this context, we will extend our studies to internal allenes and investigate their behavior in novel cyclization reactions. For the terpene alkaloid greenwaylactam A we found that the C3 position of the indole (C3 position of the "enamine" motifd) undergoes a HFIP-assisted cationic cyclization. Blocking the C3 position of the indole allowed a switch in regioselectivity and provided access to polyavolensinol via an unprecedented N-termination. This regiodivergent synthetic platform allows for the first time selective diversification via C- or N-termination.
We have also achieved the first chemical synthesis of waixenicin A, a highly potent TRPM7 ion channel inhibitor (16 nM). A sequential elongation strategy allowed selective sidechain installation and provided access to waixenicin A and enabled a biomimetic access to xeniafaraunol A. In addition, we introduced a valuable alternative to the classic Stork-Jung vinyl silane. The paved the way to access closely related natural products as well as derivatives with deep structural modifications.
In the course of our studies, we have also found reaction conditions that allow, for the first time, methyl-initiated cyclization of a wide range of substrates. In nature, the selective transfer of a methyl group to an alkene followed by cyclization involves S-adenosylmethionine (SAM)-dependent bifunctional methyltransferase cyclase (BMC) enzymes. For more than four decades, chemists have been unable to mimic BMC enzymes in the laboratory. The developed bioinspired reaction proceeds under mild conditions, is completed in less than five fours at ambient temperature, and tolerates a variety of (hetero)arenes as terminating units. The cyclization provides a variety of previously inaccessible molecular architectures in good to excellent yields. We believe that our findings are of immense interest because 1) the method is not limited by the enzymatic substrate specificity of the two known methyltransferases, and 2) the substrate scope allows escape from the canonical isoprene pattern (C5 building blocks), thus significantly expanding the chemical space.
