2-Position base-modified analogues of adenophostin A as high-affinity agonists of the D-myo-inositol trisphosphate receptor: in vitro evaluation and molecular modeling

Adenophostin A (AdA) is a potent agonist of the d-myo-inositol 1,4,5-trisphosphate receptor (Ins(1,4,5)P3R). Various 2-aminopurine analogues of AdA were synthesized, all of which (guanophostin 5, 2,6-diaminopurinophostin 6, 2-aminopurinophostin 7, and chlorophostin 8) are more potent than 2-methoxy-N6-methyl AdA, the only benchmark of this class. The 2-amino-6-chloropurine nucleoside 11, from Vorbrüggen condensation of 2-amino-6-chloropurine with appropriately protected disaccharide, served as the advanced common precursor for all the analogues. Alcoholysis provided the precursor for 5, ammonolysis at high temperature the precursor for 6, and ammonolysis under mild conditions the precursor for synthesis of 7 and 8. For 8, the debenzylation of precursor leaving the chlorine untouched was achieved by judicious use of BCl3. The reduced potency of chlorophostin 8 and higher potency of guanophostin 5 in assays of Ca2+ release via recombinant Ins(1,4,5)P3R are in agreement with our model suggesting a cation-pi interaction between AdA and Ins(1,4,5)P3R. The similar potencies of 2,6-diaminopurinophostin (6) and 2-aminopurinophostin (7) concur with previous reports that the 6-NH2 moiety contributes negligibly to the potency of AdA. Molecular modeling of the 2-amino derivatives suggests an interaction between the carboxylate side chain of Glu505 of the receptor and the 2-NH2 of the ligand, but for 2-methoxy-N6-methyl AdA the carboxylate group of Glu505 is deflected away from the methoxy group. A helix-dipole interaction between the 1-phosphate of Ins(1,4,5)P3 and the 2'-phosphate of AdA with alpha-helix 6 of Ins(1,4,5)P3R is postulated. The results support a proposed model for high-affinity binding of AdA to Ins(1,4,5)P3R.