Probing Enzyme Sequence-Function Relationships By Deep Mutational Scanning of Two Enzyme Homologs

Despite a wealth of specific knowledge on the biochemical and biophysical properties of enzymes operating in a wide range of reactions, the fundamental questions of how the enzyme sequence renders to its function remains yet unanswered. Understanding of the enzyme sequence-function relationship can yield lucrative opportunities such as enhancing protein modelling techniques, which could bolster rational design of enzymes, aid prediction of mutation disease phenotypes based, and provide insight into evolutionary potential of viral proteins or bacterial drug resistance genes.

Two main approaches are prominent in the study of protein sequence-function relationships. One such approach is the computational search for patterns of co-evolution and conservation in proteins of common evolutionary descent, in order to infer overall patterns in interactions between different parts of the protein and the importance of given parts in its function. Our lab engages in a complementary approach, known as Deep Mutational Scanning (DMS), which aims to directly generate and characterize the effects of a comprehensive set of mutations on an enzyme. The model proteins we have chosen are VIM2 and NDM1, a pair of homologous, antibiotic degrading enzymes from the metallo-beta-lactamase(MBL) superfamily. Each gene is mutated into a library of all possible single amino acid variants and placed under functional selection in Escherichia coli. Deep sequencing of the purified libraries post-selection provides enrichment levels for each variant, which is inferred to be proportional to its level of function.

I will overlay DMS data to structural features and patterns sequence conservation to reveal the molecular basis of sequence-function relationships in VIM2 and NDM1. By comparing DMS data to computational folding predictions, I can also comment on the enzyme’s behaviors in terms of protein stability and folding dynamics. The data will also allow me to discuss the evolutionary potential of MBLs for further antibiotic resistance, by revealing the effect of every possible single mutational step away from wildtype against multiple classes of antibiotics.