The Li Synthesis of Rubriflordilactone B
Rubriflordilactone B (4) shows anti-HIV 1 activity with low cytotoxicity. As the central aromatic ring insulates the two halves of 4 from each other, Ang Li of the Shanghai Institute of Organic Chemistry envisioned (Angew. Chem. Int. Ed. 2016, 55, 6964. DOI: 10.1002/anie.201601915) initial construction of 1 and 2 in enantiomerically-pure form, with coupling to 3 and cyclization to 4 at the end of the synthesis.
The preparation of 1 began from commercial perillyl alcohol 5. Arbuzov reaction with triethyl phosphite delivered the allylic phosphonate 6, that was selectively ozonized to a keto aldehyde that was cyclized to 7. Deprotonation followed by the addition of the Mukaiyama reagent 8 gave the dienone, that was selectively reduced in a conjugate sense with L-Selectride to give the transposed enone 9, without racemization. Conjugate addition of cyanide followed by Markonikov hydration and hydrolysis led to the lactone 10. The ketone was protected by methylenation, and the lactone was α-hydroxylated following the Davis protocol. Silylation and ozonolysis completed the assembly of 11, that was converted to its enol triflate. The derived acetate was cyclized by exposure to base, and the tertiary alcohol so formed was removed reductively. Allylic bromination followed by selenide-mediated dehydrobromination completed the synthesis of the diene 1.
Several creative Japanese organic synthesis chemists, including Professor Mukaiyama, were proud of Chemistry Letters, and routinely published some of their best work there. Both the reagent 8 and the Co-mediated Markonikov hydration of the alkene first appeared in that journal. There is much else in the back issues that awaits exploration by the eager practicioner.
The preparation of 2 began with the commercial enone 12. Iodination followed by Luche reduction gave 13, that was coupled with 14, then subjected to Johnson-Claisen rearrangement, to give, after desilyation, the lactone 15.
Reduction of 15 to the aldehyde oxidation state followed by exposure to 16 led, after 1,6-conjugate addition, to the cyclic ether 2. The coupling of 2 with 1 gave 3. Originally, the authors had planned selective semi-hydrogenation of 3 to the Z triene, that could be cyclized and oxidized to 4. In the event, the reduction was difficult to reproduce. Alternatively, silylation of the alkyne cleanly delivered 17 as an inconsequential mixture of regioisomers. The product from cyclization and oxidation of 17 was readily desilylated to rubriflordilactone (4).