Substrate specificity of human glycinamide ribonucleotide transformylase

Human glycinamide ribonucleotide transformylase: active site mutants as mechanistic probes

Human glycinamide ribonucleotide transformylase (GART) (EC 2.1.2.2) is a validated target for cancer chemotherapy, but mechanistic studies of this therapeutically important enzyme are limited. Site-directed mutagenesis, initial velocity studies, pH-rate studies, and substrate binding studies have been employed to probe the role of the strictly conserved active site residues, N106, H108, and D144, and the semiconserved K170 in substrate binding and catalysis. Only two conservative substitutions, N106Q and K170R, resulted in catalytically active enzymes, and these active mutant enzymes gave pH-rate profiles and a steady-state kinetic mechanism essentially identical to those of the native enzyme. All inactive mutants were able to bind both substrates, ruling out disrupted formation of the ternary complex as the source of inactivity. Differences between human and Escherichia coli GART, previously used as a model for the human enzyme, were evident.

Carbocyclic Ribosylamines: Synthesis of 5-Substituted Carbocyclic β-Ribofuranosylamines

A synthesis of 5-substituted cyclopentylamine precursors for 5'-substituted carbocyclic nucleoside analogues was developed. We show that the stereochemistry of the OsO4-catalyzed hydroxylation of an apically brominated lactam, 7-bromo-2-azabicyclo[2.2.1]hept-5-en-3-one, can be controlled through the appropriate selection of the lactam N-H protecting group. Sterically large groups direct the hydroxylation to the exo-face of the olefin, yielding hydroxylation products that can be converted into analogues of carbocyclic ribosides. Conversely, a sterically small protecting group permits OsO4 approach from the endo-face, yielding hydroxylation products analogous to carbocyclic lyxosides. A key intermediate for carbocyclic sugar production, (1S,2S,3R, 4R,5S)-1-(tert-butyloxycarbonyl)amino-5-bromo-2,3-(dimethylmethylene)dioxy-4-hydroxymethylcyclopentane, was synthesized starting from a commercially available enantiomerically pure lactam, (1S)-(+)-2-azabicyclo[2.2.1]hept-5-en-3-one, in seven steps in an overall yield of 21%.