AbstractSite‐directed mutagenesis of key amino acids of adenylate kinase has been used to suggest a new model for the location of the AMP and ATP binding sites. Phe‐86 and Tyr‐133, which are in close contact with the inhibitor Ap5A according to previous crystallographic results, have been independently changed to tryptophan and other amino acids. The Phe‐86→Trp mutant had a 3‐ to 6‐fold change in theKmfor ATP and a 44‐fold increase in theKmfor AMP with a simultaneous loss of AMP substrate inhibition. Thus Phe‐86 is probably in close contact with bound AMP. The Tyr‐133→Trp mutant showed no large effects on enzyme kinetics and suggests that the previous assignment of Ap5A occupying natural adenosine binding sites is probably incorrect. A temperature‐sensitive Leu‐107→Gln mutant showed a 6‐fold decrease in theKmfor ATP and no effect on AMP binding, suggesting that this amino acid is near the ATP binding site.Changes in the fluorescence of single tryptophan‐containing mutant enzymes provided specific information about AMP and ATP binding. The fluorescence results are consistent with the kinetic studies, and also suggest that AMP substrate inhibition is caused by the formation of an abortive complex that