Engineering of a UDP-Glycosyltransferase for the Efficient Whole-Cell Biosynthesis of Siamenoside I in Escherichia coli.

The combination of the insufficient availability and the complex structure of siamenoside I (SI), the sweetest glucoside isolated from Siraitia grosvenorii to date, limited its use as a natural sweetener. To solve this problem, an improved biocatalyst, UGT-M2, was semi-rationally created by engineering the uridine diphosphate glycosyltransferase UGT94-289-2 from S. grosvenorii for the monoglucosylation of mogroside IIIE (MG IIIE) to SI. Subsequently, an engineered Escherichia coli cell was constructed, which combined UGT-M2 with a UDP-glucose regeneration system to circumvent the need for expensive UDP-glucose to produce SI. After optimization, high-purity SI (>96.4%) was efficiently prepared from MG IIIE at a 1 L scale with a productivity of 29.78 g/(L day) and a molar yield of 76.5% and without using exogenous UDP-glucose. This study not only developed a whole-cell approach for the preparation of SI but also provided an alternative glycosyltransferase variant for SI biosynthesis with synthetic biology in the future.

Efficient whole-cell biosynthesis of l-gulose by coupling mannitol-1-dehydrogenase with NADH oxidase.

L-Gulose is a rare aldohexose to serve as a building block for anticancer drug bleomycin and nucleoside-based antivirals. However, preparative inaccessibility and high cost have hindered its pharmaceutical application. Despite a regio- and stereo-selective enzymatic synthesis of l-gulose from d-sorbitol using a variant of NAD+-dependent mannitol-1-dehydrogenase from Apium graveolens (mMDH) was explored, low efficiency and productivity caused by NADH accumulation or insufficient amount of NAD+ limited the practical utility of this process. In this study, a stable and efficient NADH oxidase from Bacillus cereus (bcNOX) was found to be more compatible with mMDH to recycle NAD+ in E. coli cells for l-gulose biosynthesis. After a systematic optimization of the whole-cell system, efficient biosynthesis of l-gulose was achieved. Starting with 70 g/L of readily available and cheap d-sorbitol resulted in a volumetric productivity of 5.5 g/L/d. This whole-cell approach enables practical, efficient and environmentally friendly biosynthesis of l-gulose and exhibits the potential of becoming a biocatalytic strategy for various enzymatic oxidative transformations.