PROTEIN DARK ENERGY IDENTIFIES LOSS OF FUNCTION MUTATIONS FAR AWAY FROM THE ACTIVE SITES

While active sites are the traditional focus of enzymology, distant amino acid substitutions can profoundly alter activity through allosteric effects. Here, we apply the concept of "Protein Dark Energy", which quantifies evolutionary fitness costs beyond mere folding thermodynamics, to evaluate these distal mutations using the D3DistalMutation database. Our analysis reveals that allosteric mutations disrupting enzymatic activity consistently display high Dark Energy signatures across diverse families, reflecting an intrinsic evolutionary sensitivity at specific distal sites rather than isolated substitution effects. Conversely, mutations that increase activity show a bimodal Dark Energy distribution. While some are explained by increased folding stability, a significant proportion incurs high Dark Energy penalties. We propose these high-penalty variants may represent deleterious gain-of-function mutations that distort metabolic pathways (a phenomenon notably prevalent in transferases) or occupy unexplored regions of sequence space.

This work was supported by Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET Grant PIP2022-2024—11220210100704CO) and by Universidad de Buenos Aires (grant UBACyT 20020220200106BA).