ABSTRACT
Penicillin G acylase (PGA) was an important biocatalyst for enzymatic production of second-generation cephalosporin. PGA from Achromobacter xylosoxidans PX02 (AxPGA) showed relatively lower identity to EcPGA (54.9% in α subunit and 51.7% in ß subunit), which could synthesize cefamandole in the kinetically controlled N-acylation (kcNa). Semi-rational design of AxPGA and "small and smart" mutant libraries were developed with minimal screening to improve cefamandole production. A triple mutant αR141A/αF142I/ßF24G by combining the mutational sites (ßF24, αR141, and αF142) from different subunits of AxPGA showed better performance in cefamandole production, with 4.2-fold of improvement in the (kcat/Km)AD value for activated acyl donor (R)-Methyl mandelate. Meanwhile, the (kcat/Km)Ps value for cefamandole by mutant αR141A/αF142I/ßF24G was sharply dropped by 25.5 times, indicating its highly synthetic activity and extremely low hydrolysis of cefamandole. Strikingly, the triple mutant αR141A/αF142I/ßF24G could form cefamandole with a yield of 85% at an economical substrate ratio (acyl donor/nucleophile) of 1.3:1 (82% at 1.1:1), which advanced the greener and more sustainable process of cefamandole production than the wild type. Furtherly, the improved synthetic ability and lower hydrolysis of cefamandole by mutant were rationalized using molecular docking.
