Chemoenzymatic entry into the (-)-podophyllotoxin family

Podophyllotoxin is an aryl tetralin lignan isolated from the plants Podophyllum peltatum and Podophyllum embodi. Aqueous extracts of these plants were used by Native Americans and peoples of the Himalayan mountains for medicinal purposes hundreds of years ago. Although podophyllotoxin has high activity against cancer cell lines it was found to exhibit prohibitive toxicity in Phase I cancer trials. Semisynthetic derivatives of the aryl tetralin lignan ($-$)-podophyllotoxin, such as etoposide and teniposide, are important chemotherapeutic agents. Etoposide has displayed remarkable efficacy as a single agent in the treatment of small cell lung cancer. Whereas podophyllotoxin has captured the attention of organic chemists for some time, most previous synthetic approaches have been racemic. Only recently have enantioselective approaches to the natural product appeared. Philosophically, our approach differs from previous asymmetric syntheses in two fundamental ways: (1) absolute stereochemistry is introduced by means of an enzyme-catalyzed desymmetrization of a meso intermediate, and (2) ring E is introduced late in the synthesis. Conditions for the efficient enzymatic desymmetrization of a tetracyclic meso intermediate were explored. Among several meso intermediates, the diacetate proved to be the most useful as an enzyme substrate. The diacetate is readily constructed in seven steps (45% yield) from piperonal. Porcine pancreatic lipase (PPL) selectively deacetylates the (R)-acetoxymethyl arm of the diacetate to furnish monoacetate (95% ee) in 84% yield. The introduction of ring E has been achieved via Cu$\rm\sp{I}$-mediated aryl-Grignard addition to the $\alpha,\beta$-unsaturated N-acyloxazolidinone derived from this 3$\sp\prime,4\sp\prime$-(methylene dioxy) cinnamate system. The conjugate addition of PhMgBr occurs exclusively from the desired re face and is quite efficient (78%). Only three steps separate conjugate addition product from the targeted analogue of ($-$)-podophyllotoxin. The facile and stereocontrolled conversion of N-acyloxazolidinone to the final target compound in just four steps attests to the potential of this synthetic route for the generation of other ring E-modified analogues of ($-$)-podophyllotoxin