Substrate specificity and structure-activity relationships of gentamicin acetyltransferase I. The dependence of antibiotic resistance upon substrate V(max)/K(m) values

Abstract

Sixteen aminoglycoside antibiotics and derivatives were found to be substrates and nine were identified as competitive inhibitors of purified gentamicin acetyltransferase I. Among the substrates, gentamicin A, kanamycin B, tobramycin, neamine, nebramine (tobramine), and gentamine C(1a) were previously questioned on the basis of results obtained from microbiological studies or enzymatic experiments with crude bacterial extracts. Gentamicin C(1a) and sisomicin were the best substrates based on V(max)/K(m) values (>107 M-1 S-1), whereas gentamicin B1 is the poorest (5x104 M-1 S-1). S(-1)). Gentamicin C1 yields the greatest V(max)(5 μmol/min/mg) while tobramycin gives the smallest (0.5 μmol/min/mg). Neomycin is the best inhibitor K(i) = 4 x 10-7 M) and 2-deoxystreptamine is the poorest (2 x 10-3 M). The kinetic characteristic which correlates with antibiotic resistance mediated by gentamicin acetyltransferase I was found to be the V(max)/K(m) value of antibiotic substrates. These kinetic changes are correlated to structural changes by a method of evaluation designed to isolate groups of kinetically independent rate constants under nonrapid and rapid equilibrium conditions. Significant findings are: although the enzyme modifies the 3-N position of the deoxystreptamine ring (II), the minimal requirements for activity includes a purpurosamine ring (I); this requirement expresses the binding and catalytic roles of the 2' and 6' (I) amines; methylation about the 6' position reduces binding, which increases reaction velocities by shifting the rate-limiting step; hydroxylation at 3' and 4' reduces catalysis but alters the roles of the 2' and 6' groups; 1-N (II) substitutions also reduce catalysis; the best substrates contain garosamine (III), effecting primarily an increase in catalysis attributed solely to the contribution of the 3'' (III) amine; the kinetic contribution of ring III, however, is greatly influenced by the identity of ring I; four-ring aminoglycosides bind differently to the enzyme, perhaps inversely. Four enzymatic structure-activity relationships closely correlate with antimicrobial structure-activity requirements, suggesting that attempts to chemically modify existing aminoglycoside antibiotics, in an effort to combat bacterial resistance, are not likely to succeed.