Mechanistically diverse enzyme superfamilies represent sets of divergent proteins whose substrates, products and even overall functions can be substantially different. Divergent evolution of such broadly varied chemical reactions can be described by the chemistry-constrained model of enzyme evolution, in which nature re-engineers the ancestral scaffold for a variety of functions by conserving a fundamental chemical capability such as a partial reaction, while evolving variations in substrate binding, and therefore overall chemistry. This renewal proposal has three aims, which extend the progress achieved in the previous grant. 1) Investigate additional mechanistically diverse enzyme superfamilies to determine how the delivery of catalysis is constrained by the common catalytic module in each. We will also detail for each how new catalysts have arisen to perform a variety of functions. We expect the results to reveal general principles of enzyme design utilized in nature and identify specific rules applicable for functional inference and mechanistic understanding for each of the superfamilies investigated. These rules will be used to predict molecular function/functional properties of superfamily sequences. This information will be made available to the scientific community via our """"""""Structure-Function Linkage Database (SFLD)"""""""" to aid others in inference of function and to guide protein engineering/design for applications to human health. 2) Identify sequence/structural differences that discriminate subgroups/families in characterized superfamilies to achieve more precision in functional inference than can be obtained by prediction of the superfamily-common functions alone. Because members of mechanistically diverse superfamilies show high levels of misannotation in public databases, we will also use this information to identify and correct misannotated sequences in these superfamilies to the extent possible. 3) Investigate superfamilies that utilize complex co-factors to learn how such superfamilies differ from the relatively more """"""""simple"""""""" types of superfamilies we have previously studied. These studies will focus first on superfamilies that use FAD cofactors.
Showing the most recent 10 out of 71 publications
