Lerner L M
J Med Chem. 1982 Jul;25(7):825-8. doi: 10.1021/jm00349a011.
A previously reported preparation of 9-(3,5-dideoxy-beta-D-glycero-pent-4-enofuranosyl)adenine (5) was not correct. A new synthesis of 5 is described. 3,5-Dideoxy-5-iodo-1,2-O-isopropylidene-alpha-D-erythro-pentofuranose (1) and 3,5-dideoxy-5-iodo-1,2-O-isopropylidene-beta-L-threo-pentofuranose (6) were prepared as starting materials. The isopropylidene groups were exchanged for acetyl groups by acetolysis, and the resulting diacetates (2 and 7) were isopropylidene groups were coupled with 6-(benzamidochloromercuri)purine by the titanium tetrachloride method. The blocked nucleosides (3 and 8) were separated from unreacted sugars by chromatography and were treated with 1,8-diazabicyclo-[5.4.0]undec-7-ene, followed by removal of the blocking groups. The alpha and beta anomers of 9-(3,5-dideoxy-D-glycero-pent-4-enofuranosyl)adenine (4 and 5, respectively) were obtained from 3 in the ratio of 3:1. Condensation of 7 with the base gave the beta-nucleoside 5 in a higher yield; no 4 was detected. When 8 was treated with sodium benzoate in hot DMF and the blocking groups were removed, 9-(5-methyl-2-furyl)adenine (9) and 9-(3-deoxy-alpha-benzoate in hot DMF and the blocking groups were removed, 9-(5-methyl-2-furyl)adenine (9) and 9-(3-deoxy-alpha-L-threo-pentofuranosyl)adenine (10) were the products. It was demonstrated that sodium benzoate was solely responsible for formation of the 5-methyl-2-furyl ring and that the first step was formation of the 4',5'-olefin.