Metabolic engineering of Escherichia coli into a versatile glycosylation platform: production of bio-active quercetin glycosides

Metabolic Engineering of Escherichia coli for Astragalin Biosynthesis

Astragalin (kaempferol 3-O-glucoside) is used as a standard to assess the quality of Radix astragali and has exhibited a number of biological properties. In this work, we screened several UDP-dependent glycosyltransferases (UGT) for their potential as efficient biocatalysts for astragalin synthesis. The highest astragalin production with 285 mg/L was detected in the recombinant strain expressing UGT from Arabidopis thaliana (AtUGT78D2). To further improve astragalin production, an efficient UDP-glucose synthesis pathway was reconstructed in the recombinant strain by introducing sucrose permease, sucrose phosphorylase, and uridylyltransferase. On the basis of those results, a recombinant strain, BL21-II, was constructed to produce astragalin. By optimizing conversion conditions, astragalin production was increased from 570 to 1708 mg/L. The production was scaled up using a fed-batch fermentation, and maximal astragalin production was 3600 mg/L, with a specific productivity of 150 mg/L/h after 24 h incubation and a corresponding molar conversion of 91.9%, the highest yield reported to date.

Engineering Saccharomyces cerevisiae with the deletion of endogenous glucosidases for the production of flavonoid glucosides

BACKGROUND

Glycosylation of flavonoids is a promising approach to improve the pharmacokinetic properties and biological activities of flavonoids. Recently, many efforts such as enzymatic biocatalysis and the engineered Escherichia coli biotransformation have increased the production of flavonoid glucosides. However, the low yield of flavonoid glucosides can not meet the increasing demand for human medical and dietary needs. Saccharomyces cerevisiae is a generally regarded as safe (GRAS) organism that has several attractive characteristics as a metabolic engineering platform for the production of flavonoid glucosides. However, endogenous glucosidases of S. cerevisiae as a whole-cell biocatalyst reversibly hydrolyse the glucosidic bond and hinder the biosynthesis of the desired products. In this study, a model flavonoid, scutellarein, was used to exploit how to enhance the production of flavonoid glucosides in the engineered S. cerevisiae.

RESULTS

To produce flavonoid glucosides, three flavonoid glucosyltransferases (SbGTs) from Scutellaria baicalensis Georgi were successfully expressed in E. coli, and their biochemical characterizations were identified. In addition, to synthesize the flavonoid glucosides in whole-cell S. cerevisiae, SbGT34 was selected for constructing the engineering yeast. Three glucosidase genes (EXG1, SPR1, YIR007W) were knocked out using homologous integration, and the EXG1 gene was determined to be the decisive gene of S. cerevisiae in the process of hydrolysing flavonoid glucosides. To further enhance the potential glycosylation activity of S. cerevisiae, two genes encoding phosphoglucomutase and UTP-glucose-1-phosphate uridylyltransferase involved in the synthetic system of uridine diphosphate glucose were over-expressed in S. cerevisiae. Consequently, approximately 4.8 g (1.2 g/L) of scutellarein 7-O-glucoside (S7G) was produced in 4 L of medium after 54 h of incubation in a 10-L fermenter while being supplied with ~3.5 g of scutellarein.

CONCLUSIONS

The engineered yeast harbouring SbGT with a deletion of glucosidases produced more flavonoid glucosides than strains without a deletion of glucosidases. This platform without glucosidase activity could be used to modify a wide range of valued plant secondary metabolites and to explore of their biological functions using whole-cell S. cerevisiae as a biocatalyst.

Recent developments in the enzymatic O-glycosylation of flavonoids

The glycosylation of bioactive compounds, such as flavonoids, is of particular relevance, as it modulates many of their pharmacokinetic parameters. This article reviews the literature between 2010 and the end of 2015 that deals with the enzymatic O-glycosylation of this class of compounds. Enzymes of glycosyltransferase family 1 remain the biocatalysts of choice for glycodiversification of flavonoids, in spite of relatively low yields. Transfers of 14 different sugars, in addition to glucose, were reported. Several Escherichia coli strains were metabolically engineered to enable a (more efficient) synthesis of the required donor during in vivo glycosylations. For the transfer of glucose, enzymes of glycoside hydrolase families 13 and 70 were successfully assayed with several flavonoids. The number of acceptor substrates and of regiospecificities characterized so far is smaller than for glycosyltransferases. However, their glycosyl donors are much cheaper and yields are considerably higher. A few success stories of enzyme engineering were reported. These improved the catalytic efficiency as well as donor, acceptor, or product ranges. Currently, the development of appropriate high-throughput screening systems appears to be the major bottleneck for this powerful technology.