Approaches to engineering and characterizing novel biological catalysts.

摘要

Experimentally, we have considered both reductive and inductive approaches to constructing and characterizing novel biological catalysts. The former approach condenses a complex data set to simplest terms and the latter proposes analysis of a particular observation as a means of extrapolating a general solution.;From a reductionist's perspective, we have evaluated the contributions of secondary structure motifs to substrate specificities exhibited by the vertebrate and bacterial forms of the enzyme dihydrofolate reductase. Our efforts focused on engineering a bacterial reductase with enhanced specificity for folic acid. Using the kinetic and structural data available on the Escherichia coli and chicken liver reductases, we constructed mutant bacterial reductases with vertebrate secondary structure and characterized them with steady-state and pre-steady-state kinetics, circular dichroism spectroscopy and spectrofluorometric determinations of ligand dissociation constants. Although enhanced substrate specificity was observed for one mutant, all the mutants were "bacterial" in terms of catalytic behavior. Based on this work, we conclude that several mutations in the bacterial enzyme will be necessary to enhance specificity for folic acid.;The inductive approach has its origins in our understanding of the role of enzymes in facilitating chemical transformations. Investigators in the field of catalytic antibodies have demonstrated that structure of a transition-state analogue is sufficient to induce the immune system to produce novel and effective antibody catalysts. We have explored the possibility of reconstituting the analogue-specific antibody repertoire by cloning and characterizing catalytic antibody gene fragments in Escherichia coli. This research demonstrates that antibody fragments produced by and purified from bacteria retain the catalytic properties of the full length monoclonal. Moreover, our studies of phage ;Using this cloning strategy, we constructed three libraries with binding specificity for p-nitrophenylphosphonamidate, a transition-state analogue for amide hydrolysis. We demonstrate that these combinatorial libraries contain a large, diverse population of clones with high affinity for the analogue.