Mammalian phosphodiesterases (PDEs) hydrolyze cyclic nucleotides and are essential in the regulation of cAMP-mediated signal transduction. They constitute a large group of isozymes that can be categorized into families or types. Different PDE types display characteristic tissue distributions, distinctive biochemical properties and unique inhibitor profiles. These features account for the physiological properties of these enzymes and are directly relevant to the efficacy of PDE inhibitors as therapeutic agents. Sequence alignment of different PDE types shows some conservation that is likely to reflect requirements for catalytic activity. The specific regions, and the particular residues, responsible for the distinctive properties of each PDE type are not yet known, however. We previously devised a yeast expression system for analysis of mammalian PDEs and for genetic selection of drug resistant mutants of these enzymes. Deletions and point mutations were used to identify regions and residues necessary for activity and for drug binding. We now propose to examine isozyme specificity in several PDE types, making use of a modified approach that builds on the strengths of our reconstituted yeast system. We will utilize chimeric PDE enzymes to determine which residues are sufficient, within the context of a PDE, to confer the unique biochemical properties and inhibitor susceptibility of each PDE type. These data will be used to optimize the selection of drug-resistant mutants leading to the identification of specific residues that account for these properties. Experiments to examine substrate specificity are also proposed. The results will address whether isozyme specificity determinants are localized in a pattern that is conserved across divergent forms of PDEs. In addition, the information derived from these studies should be useful for PDE-centered therapeutics where distinction among isozymes is crucial. It may also serve as a model for the analysis of other complex isozyme families.
