Comprehensive engineering of novel glycosyltransferase for efficient, donor-promiscuous, and regioselective glycosylation of flavonoids.

Flavonoid O-glycosylation, catalyzed by uridine diphosphate (UDP)-glycosyltransferases, is crucial for their therapeutic efficacy. However, most UDP-glycosyltransferases encounter three major limitations: low activity, poor regioselectivity, and restricted substrate availability, hindering their pharmaceutical applications. To address these challenges, we conducted protein engineering on a previously unidentified glycosyltransferase, UGT75AJ2, which had 3',7-O-glycosylation capabilities. Our approach involved three strategies: (i) development of a tailored focused rational iterative site-specific mutagenesis strategy, augmented by virtual screening and iterative mutagenesis, to design mutant Mut4-1 (S367A/V274A/F82V/I132T) with a 128-fold enhancement in relative catalytic activity; (ii) enhancement of the enzyme's compatibility with a broader spectrum of sugar donors achieved through structural-based engineering, yielding mutant S14G/F366H/S367G and demonstrating effective utilization of diverse donors; (iii) construction of a targeted mutant library to enhance regioselectivity by active site analysis, leading to mutants with high selectivity for targeted glycosylation sites. This comprehensive study tackles predominant challenges in UDP-glycosyltransferases protein engineering, providing innovative approaches and insights that enhance the development of flavonoid glycosides.