Computational framework-guided engineering of microbial transglutaminase enables site-specific conjugation of Trastuzumab.

Microbial transglutaminase (MTGase) specifically catalyzes the conjugation between primary amine-containing molecules and glutamine residues, making it a promising enzymatic approach for producing homogeneous antibody-drug conjugates (ADCs). However, this strategy faces challenges associated with glycan-induced steric hindrance, insufficient enzyme activity, and poor thermal stability. In this study, we engineered MTGase from Streptomyces mobaraensis (MTGase-Sm) using a computational framework-guided strategy to enhance its catalytic cc for site-specific conjugation at the Q295 site of trastuzumab. A triple mutant M3-MTGase-Sm (R48K/G283S/S303G) exhibited a catalytic activity of 21.1 U/mg toward the model substrate CBZ-Gln-Gly, representing a 3.6-fold improvement over the wild-type enzyme. Furthermore, the variant demonstrated enhanced thermostability. Its half-life at 50 °C was 65 min, approximately a 1.4-fold increase over that of the wild-type MTGase-Sm. Structural and molecular dynamics simulations further revealed the underlying mechanism. These mutations regulate the local structural flexibility of the enzyme, stabilize the geometric configuration of the active center, and optimize long-range conformational dynamics, thereby effectively improving enzymatic properties. The engineered M3-MTGase-Sm can conjugate small-molecule probes (dansylcadaverine) and polyethylene glycol derivatives (mPEG-NH2) to the conserved glutamine residue Q295 of trastuzumab. Notably, it maintains improved conjugation performance even in the presence of native N-glycans at the adjacent sterically hindered N297 site. Nevertheless, this variant has not yet been validated for conjugation with cytotoxic agents. This study establishes a computation-informed framework for MTGase evolution, and the engineered MTGase demonstrates significant potential for the biosynthesis of ADCs with high specificity and uniformity.