摘要：

Affinity selection from peptide libraries is a powerful tool that has been used for determining the sequence specificities of a number of enzymes and protein binding domains, including protein kinases, src homology 2 domains, and PDZ domains. We have extended this approach to protein tyrosine phosphatases using peptide libraries containing a nonhydrolyzable phosphotyrosine analog, difluorophos-phonomethylphenylalanine. A size-exclusion method is used to separate enzyme–peptide complexes from free peptide, providing several advantages over the traditional immobilized protein affinity column approach. In addition, the feasibility of using mass spectrometric detection to quantitate peptides rapidly and reproducibly is demonstrated as an alternative to quantitation by peptide sequencing. The validity of this analysis is demonstrated by synthesizing individual peptides and comparing their affinity for enzyme with the predictions from the affinity selection process. As a model for these studies the protein tyrosine phosphatase PTP1B is used, providing additional insights into the sequence specificity of this enzyme. In particular, a selection for aromatic amino acids at the pY − 1 position (immediately N-terminal to the phosphotyrosine), as well as a broad pY + 1 selectivity, is observed in addition to the general preference for acidic residues N-terminal to the phosphotyrosine. The approach described here should prove applicable to protein tyrosine phosphatases in general as well as for the study of nonpeptidyl combinatorial libraries.

展开