Metabolic engineering of the L-phenylalanine pathway in Escherichia coli for the production of S- or R-mandelic acid

Abstract

Background: Mandelic acid (MA), an important component in pharmaceutical syntheses, is currently produced exclusively via petrochemical processes. Growing concerns over the environment and fossil energy costs have inspired a quest to develop alternative routes to MA using renewable resources. Herein we report the first direct route to optically pure MA from glucose via genetic modification of the L-phenylalanine pathway in E. coli.

Results: The introduction of hydroxymandelate synthase (HmaS) from Amycolatopsis orientalis into E. coli led to a yield of 0.092 g/L S-MA. By combined deletion of competing pathways, further optimization of S-MA production was achieved, and the yield reached 0.74 g/L within 24 h. To produce R-MA, hydroxymandelate oxidase (Hmo) from Streptomyces coelicolor and D-mandelate dehydrogenase (DMD) from Rhodotorula graminis were co-expressed in an S-MA-producing strain, and the resulting strain was capable of producing 0.68 g/L R-MA. Finally, phenylpyruvate feeding experiments suggest that HmaS is a potential bottleneck to further improvement in yields.

Conclusions: We have constructed E. coli strains that successfully accomplished the production of S- and R-MA directly from glucose. Our work provides the first example of the completely fermentative production of S- and R-MA from renewable feedstock.

Figure 1

Schematic representation of S-MA or R-MA biosynthesized from glucose in engineered E. coli. hmaS: Hydroxymandelate synthase gene from A. orientalis; hmo: Hydroxymandelate oxydase gene from S. coelicolor; dmd: D-mandelate dehydrogenase gene from R. graminis; E4P: erythrose-4-phosphate; PEP: phosphoenolpyruvate; CHOR: chorismate; HPP: hydroxyphenylpyruvate; DAHP: 3-deoxy-D-arabino-heptulosonate-7-phosphate.

Figure 2

Recombinant plasmids constructed in this study. (A) The recombinant plasmid pSUFAAQ used for S-MA synthesis. (B) The recombinant plasmid pSUFAAQSD used for R-MA synthesis.

Figure 3

Biosynthesis of S-MA in different mutants with pSUFAAQ. Different deletion mutants containing recombinant pSUFAAQ were cultivated in shake flasks for 24 h. The product S-mandelic acid, the intermediate phenylpyruvate, and the byproduct L-phenylalanine were analyzed. Data shown are means ± standard deviations, calculated from triplicate individual experiments.

Figure 4

Time profiles of cell growth, S- or R- mandelic acid production, and residual glucose and acetic acid accumulation during the culture of the engineered E. coli. (A) OD₆₀₀, S-mandelic acid produced, and (B) glucose consumed, byproduct acetic acid accumulated were monitored for BCAE with pSUFAAQ. (C) OD₆₀₀, R-mandelic acid produced, and (D) glucose consumed, byproduct acetic acid accumulated were monitored for BCAE with pSUFAAQSD. Data shown are means ± standard deviations calculated from triplicate individual experiments.

Figure 5

Biosynthesis of S-mandelic acid with phenylpyruvate supplied. Products of engineered strains BCAE with pSUFAAQ fed with various concentrations of phenylpyruvate were analyzed in shake flasks during 24 h. S-mandelic acid, L-phenylalanine, and phenylpyruvate (A), cell growth (B), glucose consumption (C), and yields of S-mandelic acid from supplied phenylpyruvate (D) are listed. Data shown are means ± standard deviations calculated from triplicate individual experiments.