Elimination of Byproduct Generation and Enhancement of 2′-Fucosyllactose Synthesis by Expressing a Novel α1,2-Fucosyltransferase in Engineered Escherichia coli

Abstract

2′-Fucosyllactose (2′-FL) is a kind of fucosylated human milk oligosaccharide (HMO), representing the most abundant oligosaccharide in breast milk. We conducted systematic studies on three canonical α1,2-fucosyltransferases (WbgL, FucT2, and WcfB) to quantify the byproducts in a lacZ- and wcaJ-deleted Escherichia coli BL21­(DE3) basic host strain. Further, we screened a highly active α1,2-fucosyltransferase from Helicobacter sp. 11S02629-2 (BKHT), which exhibits high in vivo 2′-FL productivity without the formation of byproducts difucosyl lactose (DFL) and 3-FL. The maximum 2′-FL titer and yield reached 11.13 g/L and 0.98 mol/mol of lactose, respectively, in shake-flask cultivation, both approaching the theoretical maximum value. In a 5 L fed-batch cultivation, the maximum 2′-FL titer reached 94.7 g/L extracellularly with a yield of 0.98 mol of 2′-FL/mol of lactose and productivity of 1.14 g L–1 h–1. Our reported 2′-FL yield is the highest from lactose reported to date.

An Overview of Sugar Nucleotide-Dependent Glycosyltransferases for Human Milk Oligosaccharide Synthesis

“…Some glycosyltransferases exhibit flexible substrate specificity and regioselectivity and thus are prone to generating some byproducts. As mentioned above, some α1,2-FucTs generated DFL and/or 3-FL when producing 2′-FL, , in vivo production of LNFP I and Fucα1–2Galβ1–4GlcNAcβ1–3Galβ1–4Glc might be accompanied by the generation of considerable amount of 2′-FL; α1,3- and α1,3/4-FucTs might fucosylate both the GlcNAc in LacNAc unit and Glc in the terminal Lac; and SiaTs might produce some byproducts because of their multifunctional properties . Screening of specific glycosyltransferases is an important research topic to solve these problems.…”

“…Recently, Zhu et al used the high-performance anion-exchange chromatography-pulsed amperometric detection (HPAEC−PAD) to simultaneously detect these three FLs and proved that H. pylori FucT2 and E. coli WbgL showed α1,2-FucT activity and produced DFL and/or 3-FL as minor products. 68 It is possible that previous studies might ignore the accurate detection of byproducts, so there actually might be more α1,2-FucTs possessing α1,3-FucT activity in the reported α1,2-FucTs.…”

“…However, recently, the additional α1,3-FucT activity of these typical α1,2-FucTs has received much attention, which results in the production of byproducts during microbial synthesis of 2′-FL, especially difucosyllactose (DFL) . For examples, E. coli WbgL-based engineered strains produced DFL and/or 3-fucosyllactose (3-FL) as the minor products during 2′-FL synthesis, , and H. pylori FucT2 also produced DFL as a minor product in both engineered E. coli and Saccharomyces cerevisiae . However, B. fragilis WcfB interestingly did not produce DFL during 2′-FL synthesis, but its 2′-FL synthesis ability was relatively low, with the titer of 15.4 g/L in a fed-batch cultivation .…”

“…The reported bacterial α1,2-FucTs generally show 20%− 40% amino acid sequence identity with one another. 68 The overall identity is relatively low, however, all the bacterial α1,2-FucTs interestingly display four motifs with highly conserved sequences, in which motifs I, II, and III are close to the Cterminus and motif IV is located in the N-terminus (Figure 3). 71 The motif I, HxRRxD, is also conserved in the mammalian α1,6-FucT, whose crystal structure indicates its role as donor GDP-Fuc binding domain.…”

“…One possible reason is that the effective separation and accurate detection of 2′-FL, 3-FL, and DFL are relatively difficult, and in particular, the two isomers, 2′-FL and 3-FL, have the same molecular mass and similar structure and are difficult to effectively separate and accurately detect by regular liquid chromatography and/or mass spectrometry. Recently, Zhu et al used the high-performance anion-exchange chromatography-pulsed amperometric detection (HPAEC–PAD) to simultaneously detect these three FLs and proved that H. pylori FucT2 and E. coli WbgL showed α1,2-FucT activity and produced DFL and/or 3-FL as minor products . It is possible that previous studies might ignore the accurate detection of byproducts, so there actually might be more α1,2-FucTs possessing α1,3-FucT activity in the reported α1,2-FucTs.…”