[Protein engineering of PgsA combined with Bayesian optimization drives efficient poly-γ-glutamic acid synthesis by...

Poly-γ-glutamic acid (γ-PGA) is a biopolymer polymerized from l-glutamic acid (l-Glu) and/or d-glutamic acid monomers, exhibiting broad application prospects in pharmaceuticals, cosmetics, and other fields. The low yields remain a key factor limiting large-scale production of γ-PGA in heterologous expression systems. To enhance the biosynthetic efficiency, this study employed Corynebacterium glutamicum as the chassis cell and utilized the γ-PGA synthase PgsBCA from Bacillus licheniformis to catalyze γ-PGA synthesis. First, a polycistron expression system was constructed, increasing the γ-PGA yield by 20.6% compared with the monocistron system. After site-directed and saturation mutagenesis of PgsA, the mutant K76C was identified, increasing the γ-PGA yield by 75.3%. On this basis, a Bayesian model was employed to optimize seven medium components including glucose and urea. The optimized medium achieved a γ-PGA yield of 14.52 g/L, which represented a 28.6% increase over that in the initial medium. Finally, the recombinant strain was scaled up in a 5 L fermenter. After 48 hours of fermentation, the γ-PGA yield reached 64.13 g/L, with a glucose conversion rate of 0.52 g/g and the l-Glu content of 100%. This study developed a high-yielding γ-PGA-producing engineered strain and established an efficient fermentation process, providing a novel strategy for the efficient production of γ-PGA.