Metabolic engineering of Agrobacterium sp. strain ATCC 31749 for production of an alpha-Gal epitope

Abstract

Background: Oligosaccharides containing a terminal Gal-alpha1,3-Gal moiety are collectively known as alpha-Gal epitopes. alpha-Gal epitopes are integral components of several medical treatments under development, including flu and HIV vaccines as well as cancer treatments. The difficulty associated with synthesizing the alpha-Gal epitope hinders the development and application of these treatments due to the limited availability and high cost of the alpha-Gal epitope. This work illustrates the development of a whole-cell biocatalyst for synthesizing the alpha-Gal epitope, Gal-alpha1,3-Lac.

Results: Agrobacterium sp. ATCC 31749 was engineered to produce Gal-alpha1,3-Lac by the introduction of a UDP-galactose 4'-epimerase:alpha1,3-galactosyltransferase fusion enzyme. The engineered Agrobacterium synthesized 0.4 g/L of the alpha-Gal epitope. Additional metabolic engineering efforts addressed the factors limiting alpha-Gal epitope production, namely the availability of the two substrates, lactose and UDP-glucose. Through expression of a lactose permease, the intracellular lactose concentration increased by 60 to 110%, subsequently leading to an improvement in Gal-alpha1,3-Lac production. Knockout of the curdlan synthase gene increased UDP-glucose availability by eliminating the consumption of UDP-glucose for synthesis of the curdlan polysaccharide. With these additional engineering efforts, the final engineered strain synthesized approximately 1 g/L of Gal-alpha1,3-Lac.

Conclusions: The Agrobacterium biocatalyst developed in this work synthesizes gram-scale quantities of alpha-Gal epitope and does not require expensive cofactors or permeabilization, making it a useful biocatalyst for industrial production of the alpha-Gal epitope. Furthermore, the engineered Agrobacterium, with increased lactose uptake and improved UDP-glucose availability, is a promising host for the production of other medically-relevant oligosaccharides.

Figure 1

Metabolic pathway for Gal-α1,3-Lac synthesis. Metabolic pathways of ATCC 31749 for Gal-α1,3-Lac synthesis. Cofactors are shown in bold and enzymes are in bold italics. The double dashed lines indicate the cell membrane, and the dashed arrows represent transport reactions.

Figure 2

Plasmids constructed in this study. A: pBQET, containing the galE:α1,3-galT fusion enzyme for Gal-α1,3-Lac synthesis; B: pBQETY, containing the galE:α1,3-galT fusion enzyme and lacY for Gal-α1,3-Lac synthesis and lactose transport; C: pBScrdShG, containing homologous regions of crdS interrupted by a gentamicin resistance cassette for crdS knockout.

Figure 3

Gal-α1,3-Lac synthesis by ATCC 31749/pBQET. Synthesis of Gal-α1,3-Lac by ATCC 31749/pBQ (control) and ATCC 31749/pBQET. Data points are averages of three independent experiments with the standard deviation indicated by error bars.

Figure 4

Comparison of Gal-α1,3-Lac synthesis by ATCC 31749/pBQET and ATCC 31749/pBQETY. Synthesis of Gal-α1,3-Lac after 150 hours by ATCC 31749/pBQET and ATCC 31749/pBQETY without and with rifampicin. Data are averages of three independent experiments with the standard deviation indicated by error bars.

Figure 5

Aniline blue staining of curdlan. Aniline blue staining of curdlan production in wild-type and crdS mutant strains. A: ATCC 31749; B: ATCC 31749ΔcrdS colony 8; C: LTU265; D: ATCC 31749ΔcrdS colony 14.

Figure 6

Gal-α1,3-Lac synthesis by the engineered crdS mutants. Synthesis of Gal-α1,3-Lac by ATCC 31749ΔcrdS/pBQET and ATCC 31749ΔcrdS/pBQETY. Data points are averages of three independent experiments with the standard deviation indicated by error bars.