Formamide-based production of amines by metabolically engineering Corynebacterium glutamicum

Abstract

Formamide is rarely used as nitrogen source by microorganisms. Therefore, formamide and formamidase have been used as protection system to allow for growth under non-sterile conditions and for non-sterile production of acetoin, a product lacking nitrogen. Here, we equipped Corynebacterium glutamicum, a renowned workhorse for industrial amino acid production for 60 years, with formamidase from Helicobacter pylori 26695, enabling growth with formamide as sole nitrogen source. Thereupon, the formamide/formamidase system was exploited for efficient formamide-based production of the nitrogenous compounds L-glutamate, L-lysine, N-methylphenylalanine, and dipicolinic acid by transfer of the formamide/formamidase system to established producer strains. Stable isotope labeling verified the incorporation of nitrogen from formamide into biomass and the representative product L-lysine. Moreover, we showed ammonium leakage during formamidase-based access of formamide to be exploitable to support growth of formamidase-deficient C. glutamicum in co-cultivation and demonstrated that efficient utilization of formamide as sole nitrogen source benefitted from overexpression of formate dehydrogenase. KEY POINTS: • C. glutamicum was engineered to access formamide. • Formamide-based production of nitrogenous compounds was established. • Nitrogen cross-feeding supported growth of a formamidase-negative strain.

Keywords: Amine production; Co-cultivation; Corynebacterium glutamicum; Formamidase; Formamide.

Fig. 1

Influence of formamide and formate on the growth of C. glutamicum strains. Biomass (dark blue circles), µ_max (light blue triangles), and lag phase (purple squares) of strains WT-EV (open symbols) and FORM (closed symbols), cultivated in CgXII minimal medium (a, b, d, e) or N-CgXII (c) supplemented with 40 g L⁻¹ glucose and either 0–160 mM formamide (a, b, c) or 0–160 mM sodium formate (d, e) in a BioLector microcultivation system for 96 h. Values represent means with standard deviations of triplicate cultivations

Fig. 2

AmiF activity in crude extracts (a), ¹⁵N labeling of L-lysine from ¹⁵N-labeled formamide (b), and growth (c) of formamidase-expressing strain FORM (blue) and WT-EV (grey) with formamide as sole nitrogen source. Cells were grown in LB over night for crude extract preparation (a). Cells were grown in CgXII minimal medium containing 468 mM unlabeled (14N) N in form of urea and (NH₄)₂SO₄, supplemented with 60 mM ¹⁵N labeled formamide (15N) and 40 g L⁻¹ glucose, in a BioLector microcultivation system for 24 h. The patterns depict the fractions of 0 (grey), 1 (red), or 2 (purple) labeled nitrogen atoms per molecule of lysine in the biomass. Standard deviations refer to triplicate cultivations. For growth assessment, cells were grown in N-CgXII minimal medium, supplemented with 40 mM formamide for 28 h (c). Values represent means with standard deviations of triplicate measurements or cultivations. *n.d., not detectable (< 0.4 U mg.⁻¹)

Fig. 3

¹⁵N labeling of L-lysine, L-alanine, L-serine, L-proline, and L-tyrosine in strain FORM. Cells were grown in N-CgXII, supplemented with 60 mM ¹⁵N-labeled (NH₄)₂SO₄ or formamide and 40 g L⁻¹ glucose, in a BioLector microcultivation system for 24 h. The patterns depict the fractions of 0 (grey), 1 (red), or 2 (purple) labeled nitrogen atoms per molecule of lysine in the biomass. Standard deviations refer to triplicate cultivations

Fig. 4

¹⁵N labeling of lysine in WT-EV and FORM from ¹⁵N-labeled ammonium sulfate or formamide. Cells were grown in N-CgXII, supplemented with 60 mM unlabeled (¹14N) or labeled (15N) (NH₄)₂SO₄ or formamide (a), or with 30 mM labeled or unlabeled (NH₄)₂SO₄ and formamide, respectively (b), and 40 g L⁻¹ glucose, in a BioLector microcultivation system for 24 h. The patterns depict the fractions of 0 (grey), 1 (red), or 2 (purple) labeled nitrogen atoms per molecule of lysine in the biomass. Standard deviations refer to triplicate cultivations. *n.g., no growth with formamide

Fig. 5

Free ammonium in supernatants of cells of WT-crimson (a) and FORM-gfp (b). Cells were either grown separately (a, b) or in co-culture inoculated with 70% WT-crimson and 30% FORM-gfp (c), supplemented with 60 mM N in form of urea and (NH₄)₂SO₄ (grey) or with 60 mM formamide (blue) and 40 g L⁻¹ glucose as carbon source, for 24 h. Values represent means of triplicate measurements with standard deviations. *n.g., no growth with formamide

Fig. 6

Co-cultures of strains WT-crimson (red) and FORM-gfp (green) in varied inoculum ratios. To test if FORM-gfp can provide nitrogen from formamide for growth of amiF-deficient WT-crimson, cells were grown with 60 mM formamide as sole nitrogen source in N-CgXII, whereas a control cultivation in CgXII minimal medium supplemented with 60 mM N in form of urea and (NH₄)₂SO₄ was also made (grey background). 40 g L^-1 glucose was added as carbon source. The inoculum contained the indicated percentage of FORM-gfp with the rest to 100% being WT-crimson cells. Biomass formation (black diamond) and culture compositions (stacked red and green bars) were determined after 0, 8, and 24 h by OD₆₀₀ and FACS analysis. Values represent means with standard deviations of triplicate cultivations

Fig. 7

Biomass formation (a), maximal growth rates µ_max (b), and lag phases (c) of FORM (blue) and formate dehydrogenase overexpressing strains FORM-Fdh_Cg (green) and FORM-Fdh_Ps (orange). Strains were cultivated in N-CgXII, supplemented with 40 g L⁻¹ glucose and 60, 120, or 160 mM formamide in a BioLector microcultivation system for 96 h. Values represent means with standard deviations of triplicate cultivations

Fig. 8

¹⁵N labelling of lysine in biomass and supernatants of strains Lys and Lys-FORM. Cells were grown in N-CgXII, supplemented with 60 mM unlabeled (14N) or labeled (15N) (NH₄)₂SO₄ or formamide, using 40 g L⁻¹ glucose, in a BioLector microcultivation system for 24 h. The patterns depict the fractions of 0 (grey), 1 (red), or 2 (purple) of labeled nitrogen atoms per molecule of lysine in the biomass (filled bars) or the supernatant (dashed bars). Values represent means with standard deviations of triplicate cultivations. *n.g., no growth with formamide

Fig. 9

Biomass formation (a), maximal growth rates µ_max (b), lag phases (c), and L-lysine titers (d) for strains Lys-FORM (blue), Fdh-overexpressing Lys-FORM-Fdh_Ps (orange), and Lys-FORM-Fdh_Cg (green). Strains were cultivated in N-CgXII, supplemented with 60 or 120 mM nitrogen in form of urea and (NH₄)₂SO₄ (filled bars) or formamide (dashed bars) and 40 g L⁻¹ glucose, in a BioLector microcultivation system for 96 h. Values represent means with standard deviations of triplicate cultivations