It has been assumed that the structure of a single inhibitor complex i
s sufficient to define the available subsites of an enzyme that has a
unique binding site and a uniquely defined mode for ligand binding - t
he specificity for these subsites can thus be probed by kinetic experi
ments. Elastase is an enzyme for which these traditional assumptions,
which underlie such structural and kinetic studies, do not hold. Three
new crystal structures of elastase complexed to chemically similar in
hibitors with similar binding affinities reveal a diversity of binding
modes as well as two new subsites on elastase. The existence of multi
ple binding sites and different binding modes for such similar inhibit
ors indicates that researchers must proceed with caution when using ki
netics to map out protein subsites.