Azepinomycin

Guanase is an important enzyme of the purine salvage pathway of nucleic acid metabolism and its inhibition has beneficial implications in viral, bacterial, and cancer therapy. The work described herein is based on a hypothesis that azepinomycin, a heterocyclic natural product and a purported transition state analog inhibitor of guanase, does not represent the true transition state of the enzyme-catalyzed reaction as closely as does iso-azepinomycin, wherein the 6-hydroxy group of azepinomycin has been translocated to the 5-position. Based on this hypothesis, and assuming that iso-azepinomycin would bind to guanase at the same active site as azepinomycin, several analogs of iso-azepinomycin were designed and successfully synthesized in order to gain a preliminary understanding of the hydrophobic and hydrophilic sites surrounding the guanase binding site of the ligand. Specifically, the analogs were designed to explore the hydrophobic pockets, if any, in the vicinity of N1, N3, and N4 nitrogen atoms as well as O(5) oxygen atom of iso-azepinomycin. Biochemical inhibition studies of these analogs were performed using a mammalian guanase. Our results indicate that (1) increasing the hydrophobicity near O(5) results in a negative effect, (2) translocating the hydrophobicity from N3 to N1 also results in decreased inhibition, (3) increasing the hydrophobicity near N3 or N4 produces significant enhancement of inhibition, (4) increasing the hydrophobicity at either N3 or N4 with a simultaneous increase in hydrophobicity at O(5) considerably diminishes any gain in inhibition made by solely enhancing hydrophobicity at N3 or N4, and (5) finally, increasing the hydrophilic character near N3 has also a deleterious effect on inhibition. The most potent compound in the series has a Ki value of 8.0±1.5μM against rabbit liver guanase.

We report an efficient, atom economical general acid-base catalyzed one-step multi-gram synthesis of azepinomycin from commercially available compounds in water. We propose that the described pH-dependent Amadori rearrangement, which couples an amino-imidazole and simple sugar, is of importance as a potential step toward predisposed purine nucleotide synthesis at the origins of life.

In our long and broad program to explore structure-activity relationships of the natural product azepinomycin and its analogues for inhibition of guanase, an important enzyme of purine salvage pathway of nucleic acid metabolism, it became necessary to investigate if the nucleoside analogues of the heterocycle azepinomycin, which are likely to be formed in vivo, would be more or less potent than the parent heterocycle. To this end, we have resynthesized both azepinomycin (1) and its two diastereomeric nucleoside analogues (2 and 3), employing a modified, more efficient procedure, and have biochemically screened all three compounds against a mammalian guanase. Our results indicate that the natural product is at least 200 times more potent toward inhibition of guanase as compared with its nucleoside analogues, with the observed K(i) of azepinomycin (1) against the rabbit liver guanase=2.5 (±0.6)×10(-6) M, while K(i) of Compound 2=1.19 (±0.02)×10(-4) M and that of Compound 3=1.29 (±0.03)×10(-4) M. It is also to be noted that while IC(50) value of azepinomycin against guanase in cell culture has long been reported, no inhibition studies nor K(i) against a pure mammalian enzyme have ever been documented. In addition, we have, for the first time, determined the absolute stereochemistry of the 6-OH group of 2 and 3 using conformational analysis coupled with 2-D (1)H NMR NOESY.