Efforts in changing protein or pathway function would be greatly aided by design principles that inform how protein sequence and structure dictate function. To elucidate these design principles, we have characterized mechanistically diverse enzyme superfamilies; groups of enzymes that share a conserved architecture of catalytic residues (or "catalytic module") that provide a conserved chemical capability used in each member's overall function. The work described in this thesis focuses on (1) how to use information from enzymes superfamilies to guide protein and pathway engineering and (2) how to detect new protein or pathway functions using selections for function.
Chapter 1 describes initial work characterizing the contribution of catalytic module residues to overall function. The functional consequences of catalytic module substitution were determined for chloromuconate lactonizing enzyme (MLE II), a highly representative member of the enolase superfamily based on sequence similarity searches.
As protein engineering is frequently dependent on the stability of mutant variants, we also wanted to characterize the stability-for-function trade-off for conserved catalytic module residues in order to estimate the stability cost of exporting the module for novel function. Chapter 2 describes the functional, stability and structural characterization of alanine-substitution mutants of o-succinylbenzoate synthase (OSBS), a member of the enolase superfamily. In this work, we show that residues that are highly conserved for function across the superfamily also contribute the most to destabilization of the protein, in agreement with past work by other groups.
Although the design and engineering of novel enzyme variants or pathways depends on sound practice in the selection of starting templates or components, high-throughput methods for functional detection are also necessary in order to identify successful outcomes. To this end, Chapter 3 describes the characterization of a selection for glutamate racemase activity. In this work, we describe methods for maximizing recovery of weak variants (typical of initial engineering mutant variants) and describe how the enrichment and recovery of mutants changes as stringency is altered. This selection was used as a foundation for the work in Chapter 4, in which superfamily functions were paired to develop a novel selection for function.
