One-step process for production of N-methylated amino acids from sugars and methylamine using recombinant Corynebacterium glutamicum as biocatalyst

Abstract

N-methylated amino acids are found in Nature in various biological compounds. N-methylation of amino acids has been shown to improve pharmacokinetic properties of peptide drugs due to conformational changes, improved proteolytic stability and/or higher lipophilicity. Due to these characteristics N-methylated amino acids received increasing interest by the pharmaceutical industry. Syntheses of N-methylated amino acids by chemical and biocatalytic approaches are known, but often show incomplete stereoselectivity, low yields or expensive co-factor regeneration. So far a one-step fermentative process from sugars has not yet been described. Here, a one-step conversion of sugars and methylamine to the N-methylated amino acid N-methyl-L-alanine was developed. A whole-cell biocatalyst was derived from a pyruvate overproducing C. glutamicum strain by heterologous expression of the N-methyl-L-amino acid dehydrogenase gene from Pseudomonas putida. As proof-of-concept, N-methyl-L-alanine titers of 31.7 g L^-1 with a yield of 0.71 g per g glucose were achieved in fed-batch cultivation. The C. glutamicum strain producing this imine reductase enzyme was engineered further to extend this green chemistry route to production of N-methyl-L-alanine from alternative feed stocks such as starch or the lignocellulosic sugars xylose and arabinose.

Figure 1

Schematic overview of the reaction catalyzed by DpkA (A) and its integration into the central carbon metabolism in C. glutamicum NMeAla1 (B). The gene deletions for improved pyruvate production are shown by black arrows with red double bars: deletion of aceE (encoding PDHE1p, the E1p subunit of the PDHC) and pqo (encoding pyruvate-quinone oxidoreductase, PQO) and both genes coding both major enzymes for l-alanine supply by pyruvate aminotransferases (alaT and avtA, encoding the alanine aminotransferase AlaT and the valine-pyruvate aminotransferase AvtA, respectively) were deleted. In addition, the acetohydroxyacid synthase (AHAS) activity was downregulated by deletion of the C-terminal part of ilvN (small subunit of AHAS) shown by red dashed arrow. Enzymes highlighted by red background indicate missing or down regulated enzymes. The thick arrow displays the NMeAla formation by heterologous expressed dpkA from P. putida KT2440 coding for the N-methylated amino acid dehydrogenase DpkA (green shadowed Enzyme).

Figure 2

Growth rates of C. glutamicum wild type in the presence of varying concentrations of MMA or NMeAla. C. glutamicum wild type was grown in presence of increasing MMA (0.05 m to 1.5 m) or NMeAla (0.05 m to 0.25 m) concentrations and specific growth rates were determined. Half maximal growth rates were obtained by extrapolation.

Figure 3

NMeAla, l-alanine, l-valine and pyruvate production data (A) and carbon balance (B) of C. glutamicum strain NMeAla1 under different conditions. Cells were cultivated in minimal medium CGXII containing 30 g L⁻¹ or 20 g L⁻¹ glucose and 16.6 g L⁻¹ potassium acetate, 2 mm l-Ala and 1 mm IPTG for induction of gene expression. The nitrogen amount of the minimal medium was reduced to 50% or 10% respectively, the glucose and MMA amount were optimized to finally 20 g L⁻¹ glucose and 10.9 g L⁻¹ MMA. The culture supernatants were harvested after incubation for 72 h and analyzed by HPLC. (A) Concentrations are given as means with standard deviation of three replicates. n.d.: not detected. (B) To assess the fate of carbon from glucose and acetate as substrates their concentrations in gram carbon per liter is plotted. The gram carbon per liter concentrations of biomass formed (green) and of the formed products l-alanine (blue), l-valine (black), pyruvate (grey), and NMeAla (red) are plotted. For NMeAla, the carbon derived from MMA was not considered. The gram carbon per liter concentrations of CO₂ and unknown byproducts are depicted in open columns. Amines < 0.1 g L⁻¹ and carbohydrates < 0.5 g L⁻¹ were not considered.

Figure 4

Fed-batch cultivation with C. glutamicum NMeAla1 in minimal medium supplemented with potassium acetate and glucose as carbon and energy sources. A fermenter with an initial start volume of 4 L was used. First feed phase (potassium acetate) was coupled to the rDOS value. After 22 h the second feed phase was started by the initial addition of 162 mL of a glucose/MMA solution followed by a linear feed of 12.3 mL h⁻¹. The biomass formation (black open squares), concentrations of NMeAla (red circles), l-alanine (blue squares), pyruvate (grey squares), acetate (green filled triangles) and glucose (green open triangles) were depicted. The volume of both feeds is shown as black lines. All depicted concentrations and the biomass formation was related to the initial volume.

Figure 5

Production of NMeAla from alternative carbon sources. The CGXII minimal medium with 16.6 g L⁻¹ potassium acetate contained 30 g L⁻¹ starch for cultivation production experiments using C. glutamicum strain NMeAla1(pECXT99A-amyA), 30 g L⁻¹ arabinose using C. glutamicum strain NMeAla1(pECXT99A-araBAD) and 30 g L⁻¹ xylose using C. glutamicum strain NMeAla1(pEKEx3-xylAB). Concentrations were determined after 72 h and are given as means with standard deviations of three replicates.