Based on the potent anticancer activity of 6'-fluorocyclopentenyl-cytosine 2b in phase IIa clinical trials for the treatment of gemcitabine-resistant pancreatic cancer, we carried out a systematic structure-activity relationship study of 6'-fluorocyclopentenyl-pyrimidines 3a-i and -purines 3j-o to discover novel anticancer agents. We also synthesized the phosphoramidate prodrug 3p of adenine derivative 1b to determine if the anticancer activity depended on the inhibition of DNA and/or RNA polymerase in cancer cells and/or on the inhibition of S-adenosylhomocysteine (SAH) hydrolase. All of the synthesized pyrimidine nucleosides exhibited much less potent anticancer activity in vitro than the cytosine derivative 2b, acting as RNA and/or DNA polymerase inhibitor, indicating that they could not be efficiently converted to their triphosphates for anticancer activity. Among all the synthesized purine nucleosides, adenine derivative 1b and N6-methyladenine derivative 3k showed potent anticancer activity, showing equipotent inhibitory activity as the positive control, neplanocin A (1a) or Ara-C. However, the phosphoramidate prodrug 3p showed less anticancer activity than 1b, indicating that it did not act as a RNA and/or DNA polymerase inhibitor like 2b. This result also demonstrates that the anticancer activity of 1b largely depends on the inhibition of histone methyltransferase, resulting from strong inhibition of SAH hydrolase. The deamination of the N6-amino group, the addition of the bulky alkyl group at the N6-amino group, or the introduction of the amino group at the C2 position almost abolished the anticancer activity.
Keywords: Anticancer; Fluorination; Fluorocyclopentenyl nucleoside; Phosphorylation; Prodrug; Structure-activity relationship.
Copyright © 2018 Elsevier Masson SAS. All rights reserved.