Small (24-35 amino acid residues) peptides that catalyze carbon-carbon bond transformations including aldol, retro-aldol, and Michael reactions in aqueous buffer via an enamine mechanism have been developed. Peptide phage libraries were created by appending six randomized amino acid residues to the C-terminus or to the N-terminus of an 18-mer R-helix peptide containing lysine residues. Reactionbased selection with 1,3-diketones was performed to trap the amino groups of reactive lysine residues that were necessary for the catalysis via an enamine mechanism by formation of stable enaminones. The selected 24-mer peptides catalyzed the reactions with improved activities. The improved activities were correlated with improved folded states of the peptides. The catalyst was then improved with respect to substrate specificity by appending a phage display-derived substrate-binding module. The resulting 35-mer peptide functioned with a significant proportion of the catalytic proficiency of larger protein catalysts. These results indicate that small designer enzymes with good rate acceleration and excellent substrate specificity can be created by combination of design and reaction-based selection from libraries.Generation of enzyme-like small designer peptide catalysts that achieve the proficiency of natural enzymes with desired substrate specificity is a challenging task. Designer protein catalysts have been prepared by engineering naturally occurring enzymes (1-4) and by using antibody libraries (i.e., immune diversity) and selection with designed small molecule compounds (5-7). For experimental purposes, small peptides are more useful and convenient than large proteins, as long as they provide the same function, because small peptides can be more easily prepared and their characterization is simpler. However, fundamental questions remain to be answered: Can small peptides attain the catalytic efficiency of larger protein catalysts, and how can these peptide catalysts best be developed? The folded states of enzymes are key to their catalytic activity since structure can be used productively to modify the chemical reactivity of amino acid side chains. Small peptides are often limited in their potential to catalyze reactions because of the limited ability to adopt well-defined structures. Small peptides have also demonstrated limited specificity for small substrate molecules.For small catalytic peptides (<50 amino acid residues) that function in aqueous buffer, rational design has been used for the creation of catalysts of the decarboxylation of oxaloacetic acid (8-12), ester hydrolysis (13-15), and transesterifications (13). Rational design, however, is limited in throughput. A combination of rational design and librarybased methods with reaction-based selection, which has been used for the development of antibody catalysts (7), may be an attractive approach to small peptide catalyst development (16,17). Here we have explored strategies for the creation of small designer enzymes. We have developed small peptides...