Optimizing non-natural protein functions with protein engineering. Evolution of a cephalosporin C acylase.

Semi-synthetic cephalosporins are synthesized starting from the 7-amino cephalosporanic acid (7-ACA) nucleus obtained from the natural antibiotic cephalosporin C (CephC). In recent years, a single-step enzymatic process in which CephC is directly converted into 7-ACA by a cephalosporin C acylase (CA) has attracted industrial interest because of the prospects of simplifying the process and reducing costs. CAs are members of the glutaryl acylase family that should use CephC as preferred substrate; however, known natural glutaryl acylases show very low activity on this antibiotic. At the host laboratory ("The Protein Factory" Research Center) in the past years the catalytic efficiency on CephC of a glutaryl acylase from Pseudomonas N176 (named VAC) was enhanced by a protein engineering approach, and the VAC crystal structure was recently solved, thus providing insight into the substrate binding and catalytic activity of CAs. However, the properties of the evolved enzymes are not sufficient to encourage 7-ACA manufacturers to shift to single-step CephC enzymatic conversion. By a combination of structural knowledge, semi-rational design, computational approaches and evolution analysis we isolated VAC variants with an altered substrate specificity (i.e. with a > 11,000-fold increase in specificity constant for CephC versus glutaryl-7-amino cephalosporanic acid, compared to wild-type) and with the highest kinetic efficiency so far obtained for a CA. This approach allowed the isolation of VAC variants suitable for the industrial application of the mono-step CephC conversion process. Indeed, taking advantage of the availability of a number of VAC variants with different kinetic properties, we setup a one-pot system in which DAAO and VAC work together to directly convert cephalosporin C into 7-ACA. The process has been optimized by identifying the most favorable operational conditions, substrate and enzymes concentrations. Under optimized conditions and the addition of further aliquots of the biocatalysts, > 98% of CephC was converted yielding 7-ACA as the main reaction product. At the 20 mL bioconversion scale, approx 81 mg of 7-ACA are produced in 41 hours from 15 mM CephC. Moreover, a study on the substrate specificity of cephalosporin derivative compounds of main industrial interest was undertaken. Molecular docking analysis allowed to deep inside the mode of binding of CephC derivatives at the active site of wild-type and different VAC variants. The ongoing molecular dynamics studies will shed light on the structure-function relationships of this class of hydrolytic enzymes as well as to identify the most suitable enzyme variant for each cephalosporin derivative.