PAI-1 as a model for mapping of protein-protein interaction sites

The identification of specific amino acid residues involved in protein-protein interaction is fundamental to understanding structure-function relationships. Therefore, we designed a procedure, supported by mathematical calculations, to perform high density mutagenesis, in order to generate a mutant library of which a limited number of random clones would suffice to exactly localize amino acids essential for a particular protein-protein interaction. Experimentally, this goal was achieved by consecutive cycles of DNA shuffling, under error prone conditions, each followed by exposure of the target protein on the surface of phages to screen and select for correctly folded, functional mutants. To validate the procedure, human plasminogen activator inhibitor 1 (PAI-1) was chosen, because its 3-D structure is known, many experimental tools are available and it may serve as a model protein for serpin structure-function studies. After 5 cycles of mutagenesis and selection for t-PA binding, analysis of 27 random clones revealed that PAI-1 mutants contained on average 9.1 amino acid substitutions distributed over 114 different positions, which were preferentially located at the surface of the protein. This small collée-tion of purified mutant PAI-1 preparations contained multiple mutants defective in recognition of 3 out of 4 tested monoclonal anti-PAI-1 antibodies (mAbs). Positional alignment of defective clones permitted assignment of single dominant amino acid residues for each mAb. The validity of these residues was unambiguously demonstrated by testing the properties of single point mutants. From the position of these amino acids in the 3-D structure and the effects of the corresponding mAbs on t-PA-PAI1 interaction, important conclusions can be drawn on this serpin-proteinase interaction.