Optimisation of engineered Escherichia coli biofilms for enzymatic biosynthesis of l-halotryptophans

Abstract

Engineered biofilms comprising a single recombinant species have demonstrated remarkable activity as novel biocatalysts for a range of applications. In this work, we focused on the biotransformation of 5-haloindole into 5-halotryptophan, a pharmaceutical intermediate, using Escherichia coli expressing a recombinant tryptophan synthase enzyme encoded by plasmid pSTB7. To optimise the reaction we compared two E. coli K-12 strains (MC4100 and MG1655) and their ompR234 mutants, which overproduce the adhesin curli (PHL644 and PHL628). The ompR234 mutation increased the quantity of biofilm in both MG1655 and MC4100 backgrounds. In all cases, no conversion of 5-haloindoles was observed using cells without the pSTB7 plasmid. Engineered biofilms of strains PHL628 pSTB7 and PHL644 pSTB7 generated more 5-halotryptophan than their corresponding planktonic cells. Flow cytometry revealed that the vast majority of cells were alive after 24 hour biotransformation reactions, both in planktonic and biofilm forms, suggesting that cell viability was not a major factor in the greater performance of biofilm reactions. Monitoring 5-haloindole depletion, 5-halotryptophan synthesis and the percentage conversion of the biotransformation reaction suggested that there were inherent differences between strains MG1655 and MC4100, and between planktonic and biofilm cells, in terms of tryptophan and indole metabolism and transport. The study has reinforced the need to thoroughly investigate bacterial physiology and make informed strain selections when developing biotransformation reactions.

Figure 1

Formation and breakdown of 5-halotryptophan in E. coli. (a) Reaction scheme for biocatalytic conversion of 5-haloindole and serine to 5-halotryptophan, catalysed by tryptophan synthase TrpBA. (b) Reaction scheme for the reverse reaction, catalysed by tryptophanase TnaA. X = F, Cl or Br.

Figure 2

Crystal Violet staining of E. coli engineered biofilms. Biofilms were generated from strains MG1655 and PHL628 (a) or MC4100 and PHL644 (b) with and without pSTB7 using the spin-down method, matured for 7 days in M63 medium and biomass was estimated using crystal violet staining.

Figure 3

Biotransformation of 5-fluoroindole to 5-fluorotryptophan using planktonic cells of four strains. Concentrations of 5-fluorotryptophan and 5-fluoroindole were measured using HPLC and percentage 5-fluorotryptophan accumulation (a), percentage 5-fluoroindole depletion (b) and the selectivity of the 5-fluoroindole to 5-fluorotryptophan reaction (c) were plotted against time. All cells contained pSTB7.

Figure 4

Biotransformation of 5-chloroindole to 5-chlorotryptophan using planktonic cells of four strains. Concentrations of 5-chlorotryptophan and 5-chloroindole were measured using HPLC and percentage 5-chlorotryptophan accumulation (a), percentage 5-chloroindole depletion (b) and the selectivity of the 5-chloroindole to 5-chlorotryptophan reaction (c) were plotted against time. All cells contained pSTB7.

Figure 5

Biotransformation of 5-fluoroindole to 5-fluorotryptophan using engineered biofilms comprising four strains. Concentrations of 5-fluorotryptophan and 5-fluoroindole were measured using HPLC and percentage 5-fluorotryptophan accumulation (a), percentage 5-fluoroindole depletion (b) and the selectivity of the 5-fluoroindole to 5-fluorotryptophan reaction (c) were plotted against time. All cells contained pSTB7.

Figure 6

Biotransformation of 5-chloroindole to 5-chlorotryptophan using engineered biofilms comprising two strains. Concentrations of 5-chlorotryptophan and 5-chloroindole were measured using HPLC and percentage 5-chlorotryptophan accumulation (a), percentage 5-chloroindole depletion (b) and the selectivity of the 5-chloroindole to 5-chlorotryptophan reaction (c) were plotted against time. All cells contained pSTB7.