Reaction rates of several reference aminoglycoside-modifying enzymes were studied at various substrate concentrations. The resulting concentration-response curves showed wide variation in threshold concentration, in curve slope, in enzyme saturation, and in substrate inhibition. Together, the curves of a defined aminoglycoside panel yielded more specific information for each individual aminoglycoside-modifying enzyme tested than did conventional substrate profiles obtained at a single substrate concentration.Enzymatic aminoglycoside resistance arises from the presence of one or more aminoglycoside phosphotransferases (APHs), aminoglycoside nucleotidyltransferases (ANTs), or aminoglycoside acetyltransferases (AACs). To determine to which group an aminoglycoside-modifying enzyme (AME) belongs, specific radioactively labeled coenzymes,]acetyl coenzyme A, are used as cosubstrate in enzymatic reactions. With regard to the aminoglycoside modification site, several specific types can be distinguished within each group. The modification site is determined by examining the chemical structure of the modified aminoglycoside. For isoenzyme typing, discriminative enzymology is required.In practice, AME identification takes place by comparison of the substrate profile of an unknown enzyme with those of reference AMEs. The substrate profiles are defined as sets of relative rates of modification of various aminoglycosides. These enzyme-specific substrate profiles are usually considered adequate for proper identification. In the past, profiles were derived from enzymatic measurements at one fixed substrate concentration (5,8,10,14). However, in a previous study (4) we observed that ANT(2") activities against various aminoglycosides appeared to be concentration dependent. Substrate inhibition occurred in several cases and may produce a falsely low index of activity against a specific substrate, even one against which the enzyme confers resistance at the lower concentrations encountered in vivo. There is, moreover, an even more common problem with the usual practice of determining substrate activity profiles. With excess substrate, i.e., concentrations at which the enzyme is saturated, the activity reflects the Vmax and is proportional to the amount of enzyme. At lower concentrations, near the K,, and below, the rate of activity should be proportional to substrate concentration. Most investigators characterize activity against a substrate profile by using a single concentration which is in excess. Thus, as pointed out by Bongaerts and Kaptijn (3) and Perlin and Lerner (15) with regard to APH(3')-II-mediated modification of amikacin, and by Bongaerts and Molendijk (4) and DeHertogh and Lerner (7) with regard to ANT(2")-mediated modification of netilmicin, excess substrate may result in greater activity against a high-* Corresponding author. t Present address: