Predictable adaptive evolution of a phage endolysin through substrate recognition optimization

Abstract

Bacteriophages release progeny by producing endolysins that degrade the bacterial cell wall. While the frequent horizontal transfer of endolysins suggests substantial evolutionary plasticity, the mechanisms by which these enzymes adapt to new phage-host contexts remain poorly understood. Here, we investigated the evolutionary dynamics and structural mechanisms governing endolysin adaptation following experimental evolution of a chimeric phage generated via heterologous endolysin exchange between phages infecting different hosts. Using replicate experimental evolution and time-resolved PacBio sequencing, we identified a dominant, highly reproducible adaptive trajectory characterized by the stepwise fixation of three key mutations. This constrained mutational order correlated with incremental gains in enzymatic activity, reflecting a rugged yet predictable fitness landscape. High-resolution structural analyses revealed that these substitutions lie exclusively outside the catalytic site; instead, they enhance substrate recognition through electrostatic tuning, optimized hydrophobic packing, and local conformational refinement, resulting in significantly higher binding affinity. While the adaptive trajectory was largely conserved, one replicate followed an alternative path, highlighting the interplay between selection and historical contingency. Adaptation was further shaped by a functional trade-off, whereby increased lytic activity on the novel host was accompanied by reduced activity on the ancestral host, consistent with antagonistic pleiotropy. Genome-wide sequencing additionally identified a compensatory mutation in a lytic transglycosylase, suggesting coordinated evolution of the broader lysis machinery. Together, these results demonstrate that endolysins evolve through reproducible adaptive walks constrained by structure, selection, and trade-offs, providing a mechanistic framework for understanding enzyme evolution and informing rational protein engineering.