J. Am. Chem. Soc. 134, 13577–13579 (2012)

Hundreds of thousands of people die each year from malaria. The natural product artemisinin has been used as a treatment alone or as part of a combination therapy, but the expense of this drug — arising from the need to extract it from a natural source (the sweet wormwood plant) — has encouraged underdosing and ultimately contributed to the development of resistant strains of the disease. Now, Chunyin Zhu and Silas Cook from Indiana University have reported a short total synthesis of (+)-artemisinin starting from inexpensive achiral starting materials.

Although several total syntheses of (+)-artemisinin have been reported, they have not been competitive on price with extraction from the natural source. The most promising current approach is semisynthetic and relies on the conversion of microbially produced (+)-artemisinic acid to the final product. A fully biosynthetic approach to (+)-artemisinin remains elusive. Zhu and Cook's aim was to design a synthesis (pictured) that relied on readily available and cheap starting materials such that their route might truly begin to compete with the semisynthetic or extraction routes.

Starting with cyclohexenone, an enantioselective copper-catalysed conjugate addition–alkylation sequence was used to set the configurations of two of the six stereocentres in the natural product while creating a compound with 11 of the 15 carbon atoms in artemisinin. The remaining carbons are introduced in a sequence involving a Shapiro reaction and an unusual Lewis-acid catalysed [4+2]-cycloaddition. The requirement for cheap starting materials in the first step meant that a Wacker oxidation was needed at this stage, but Zhu and Cook ultimately hit upon some straightforward conditions using palladium chloride and excess hydrogen peroxide.

Discovery of conditions for the final oxidative rearrangement also required some experimentation, but Zhu and Cook found success in a molybdenum-catalysed decomposition of hydrogen peroxide (which produces singlet oxygen in situ) followed by treatment with acid. The complete synthetic sequence is achieved in only five pots and, although further optimization would still be required for large-scale production, has been used to produce just over a gram of artemisinin so far.