A minoglycoside antibiotics are broad-spectrum compounds used for treatment of serious infections caused by both Grampositive and Gram-negative bacterial pathogens (1). The major mechanism of aminoglycoside resistance in Gram-positive and Gram-negative bacteria is the production of aminoglycosidemodifying enzymes, aminoglycoside phosphotransferases (APHs; also known as aminoglycoside kinases), aminoglycoside acetyltransferases, and aminoglycoside nucleotidyltransferases. These enzymes modify hydroxyl or amino groups on the antibiotics that are important for their binding to the target, the 30S subunit of the bacterial ribosome, thus significantly compromising their antimicrobial activity (2). It has been demonstrated that the APHs are capable of phosphorylating the 4-, 6-, 9-, 3=-, 2Љ-, 3Љ-, and 7Љ-hydroxyl groups of various aminoglycosides (3), and until recently, it was widely accepted that ATP is the major source of phosphate for these enzymes. Comprehensive studies of the aminoglycoside 2Љ-phosphotransferases [APH(2Љ)], enzymes which are widely distributed in Gram-positive staphylococcal (4) and enterococcal (5-7) isolates, have demonstrated that they may utilize GTP as a cosubstrate (8). Based on this knowledge, a new nomenclature for the APH(2Љ) enzymes was proposed, which reclassified the APH(2Љ)-Ia, -Ib, -Ic, and -Id enzymes as APH(2Љ)-Ia, -IIa, -IIIa, and -IVa enzymes, respectively (8). Both APH(2Љ)-Ia and APH(2Љ)-IIIa utilize exclusively GTP as a cofactor, while APH(2Љ)-IIa and APH(2Љ)-IVa can utilize both ATP and GTP (8,9). At present, it is not known whether the ability to utilize GTP for phosphorylation of aminoglycoside antibiotics is limited to the aminoglycoside 2Љ-phosphotransferases or if the phenomenon is more widespread. It is also unclear whether the ability to utilize both ATP and GTP or one of the nucleoside triphosphates (NTPs) in preference to another provides any advantage to bacteria. The concentrations of both ATP and GTP in bacterial cells are maintained in a very high, millimolar range (10, 11), implying that both NTPs are constantly available.X-ray crystallographic studies were crucial for understanding the structural requirements for the NTP profiles of the APH(2Љ) enzymes, by revealing that these kinases possess overlapping but distinct structural templates for ATP and GTP binding (12,13).Of these two templates, the ATP-binding site is located deeper in the NTP-binding pocket of the enzymes. It was further demonstrated that the inability of APH(2Љ)-IIIa [and presumably APH(2Љ)-Ia] to utilize ATP as a cosubstrate stems from the obstruction of its ATP-binding template by a bulky "gatekeeper" tyrosine residue, Tyr92 (14). To better understand the structural requirements for NTP specificity in the APH(2Љ) kinases, we attempted to convert APH(2Љ)-IIa and APH(2Љ)-IVa (both are capable of utilizing ATP and GTP as cosubstrates) into exclusively GTP-utilizing kinases by blocking their ATP-binding templates.
MATERIALS AND METHODSMutagenesis. The M85Y and F95Y substitutions were introduced ...