Directed evolution of an industrial N-acetyl-amino acid racemase
The use of stereoselective aminohydrolases (acylases) in kinetic resolutions is a commonly employed industrial route to both L- and D- α-amino acids from Nacetylated- DL-starting materials. However, a flaw in this process is the need for repeated racemisation steps of the non-desired enantiomer to achieve yields >50%. A solution to this drawback would be a dynamic kinetic resolution driven by an in situ racemisation step that would allow the yield to approach 100% A cheap and “green” catalyst for this racemisation would be an enzyme such as N-acyl amino acid racemase (NAAAR) from the actinobacteria Amycolatopsis sp. Ts-1-60. This enzyme requires no organic cofactor, is stable at high temperatures (~60°C) and importantly, shows no racemase activity toward free amino acids. Unfortunately, the activity of NAAAR with N-acetyl substrates is low and reported to suffer from inhibition above substrate concentrations of 50 mM, prohibiting its use in NAAAR/acylase coupled dynamic kinetic resolutions under industrial conditions. In an attempt to remove these limitations, directed evolution has been applied to the Amycolatopsis Ts-1-60 NAAAR to engineer a variant enzyme with increased activity towards N-acetyl substrates. An improved variant NAAAR may allow for a commercial NAAAR/acylase coupled dynamic resolution process. Directed evolution has proven to be a highly versatile and successful tool for protein engineering. However, the ability to screen for improved variants is often technically difficult or time consuming and in the case of NAAAR this is especially true as the substrate and product are simply enantiomers of an N-acetyl amino acid. To overcome this, an enantioselective genetic selection system has been employed to allow the screening of mutagenic NAAAR libraries. After several rounds of selection, a NAAAR variant (NAAAR G291D F323Y) has been isolated with increased racemase activity towards a range of synthetically useful N-acetyl substrates. This enzyme has been over-expressed, purified and its characteristics compared to the wild-type and other variants discovered during the evolution process. NAAAR G291D F323Y has been crystallised to 2.71 Å allowing a molecular basis for the increase in catalytic activity to be proposed. Coupling of this enzyme with a stereoselective acylase has been used to produce enantiopure amino acids in >99% yield, twice the inherent 50% maximum yield of acylase based resolutions. Early results suggest this NAAAR variant has the potential to be employed on a multi kilogram scale for the economical production of enantiopure L- and D- α-amino acids.