The role of intramolecular epistasis in dictating enzyme evolution

Enzyme evolution proceeds through the accumulation of mutations that rewire the intramolecular residue networks underpinning catalytic function. The context-dependence of these mutations –known as epistasis – shapes the accessibility of evolutionary trajectories. Despite its recognized importance, the prevalence of epistasis in adaptive enzyme evolution, its molecular origins, and its consequences for functional prediction remain poorly understood. This thesis develops a unified biophysical framework for deciphering the mechanistic and molecular origins of intramolecular epistasis in enzymes, and applies it to illuminate how epistasis both permits and restricts the evolution of new catalytic functions. In Chapter 2, I quantify epistasis across 41 combinatorial fitness landscapes spanning seven enzymes, demonstrating that over 94% of mutational and epistatic effects are highly idiosyncratic. This pervasive context-dependence severely distorts functional predictions along adaptive trajectories and, when examined through evolutionarily connected genotypes, exposes higher-order intramolecular network rewiring events that underpin enzyme adaptation. In Chapter 3, I expand this analysis by profiling all 64 combinations of six key phosphotriesterase mutations across nine structurally diverse substrates. Using a novel reference-based analysis pipeline incorporating error propagation and significance testing, I reveal that intramolecular network wiring varies substantially between substrates, even within the same chemical class, and that substrate-dependent epistatic signatures account for the functional trade-offs that define adaptive evolution. In Chapter 4, I identify an intrinsic, previously unrecognized source of non-specific epistasis arising from the multi-state nature of the enzyme catalytic cycle. Using kinetic cycle simulations parameterized by free energies, I demonstrate that additive perturbations to sub-state free energies produce substantial epistasis in measured kinetic parameters, provide analytical conditions for its emergence, and outline methods for its correction. In Chapter 5, I investigate how neutral mutational divergence can hinder neofunctionalization. Using methyl-parathion hydrolase as a model system, I show that three substitutions accumulating neutrally during ancestral divergence rigidify active-site loops through independent structural mechanisms, creating conformational frustration that renders five organophosphate-functionalizing mutations incompatible, providing a mechanistic basis for how historical chance entrenches enzyme evolutionary trajectories. Taken together, these findings establish intramolecular epistasis as a pervasive, mechanistically tractable force that simultaneously constrains and enables the evolution of new enzyme functions.