Chapter I. The concept and development in the field of artificial metalloenzymes (ArMs) are introduced. A general method of bioconjugation to construct ArMs via strain-promoted azide-alkyne cycloaddition (SPAAC) is described. Scaffold proteins containing a genetically encoded p-L-azidophenylalanine and catalytically active bicyclononyne-substituted metal complexes were covalently linked through SPAAC. The bioorthogonality of SPAAC allows for bioconjugation in the presence of cysteine residues in the scaffold, so no additional scaffold modification is necessary for ArM formation. The broad scope of this method with respect to both the scaffold and cofactor components was demonstrated. Catalytic study showed that a dirhodium ArM formed with this method catalyzed decomposition of diazo compounds and both SiH and olefin insertion reactions involving these compounds, but no selectivity was observed. The simplicity and modularity of the SPAAC approach should facilitate rapid optimization of the ArMs for selective catalysis. Chapter II. Rational engineering of ArMs toward selective catalysis is described. An alkyne-substituted dirhodium catalyst was linked to a prolyl oligopeptidase (POP) containing a genetically encoded p-L-azidophenylalanine residue to create an ArM that catalyzes olefin cyclopropanation. Scaffold mutagenesis based on a reported homology mode was then used to improve the enantioselectivity of this reaction, and cyclopropanation of a range of styrenes and donor–acceptor carbene precursors was accepted. Of all the mutations introduced, a histidine residue in the POP active site led to the largest improvements in both selectivity, conversion and activity, probably due to its capability to coordinate rhodium. The formed dirhodium-POP ArM also improved substrate specificity by reduced the formation of byproducts, including those resulting from the reaction of dirhodium–carbene intermediates with water. This indicated control of other water-sensitive organometallics could be possible by using solvent-sequestered POP active site. Chapter III Directed evolution efforts to improve ArM selectivity is described. A streamlined, high-throughput screening protocol for dirhodium-POP ArMs was developed. The essential step in the protocol was to scavenge excess metal cofactor without causing significant enzyme loss. Using this protocol, A POP parent mutant was submitted to iterative random mutagenesis to improve enatioselectivity in olefin cyclopropanation. Library hits giving up to 94% ee were discovered from three rounds of directed evolution. Key mutations both proximal and distal to the active site were found, which demonstrated the importance of random mutagenesis in ArM evolution. In addition, immobilization of ArMs was explored and integrated into the library screening protocol, providing an effective method to evolve ArMs for expanded scope and novel reactivity.