Proteomic analysis of Penicillin G acylases and resulting residues in semi-synthetic β-lactam antibiotics using liquid chromatography - tandem mass spectrometry.

Penicillin G acylase (PGA), as a key enzyme, is increasingly used in the commercial production of semi-synthetic β-lactam antibiotics (SSBAs). With the substitution of conventional chemical synthesis by emerging bioconversion processes, more and more PGAs fermented from different types of strains such as Escherichia coli (E. coli, ATCC 11105), Achromobacter sp. CCM 4824 and Providencia rettgeri (ATCC 31052) have been used in this kind of enzymatic processes. As an intermediate reaction catalyst, PGA protein and its presence in the final products may cause a potential risk of human allergic reaction and bring challenges for both quality and process controls. To achieve qualitative and quantitative analysis of PGAs and their residues in SSBAs, a tryptic digestion coupled with liquid chromatography - tandem mass spectrometry (LC-MS/MS) method was developed and proposed because of advantages like high selectivity and sensitivity. A suitable filter aided sample preparation (FASP) method was also used to remove matrix interference and to enrich the target PGA retained in the ultrafiltration membrane for an efficient enzymatic hydrolysis and subsequent accurate MS detection. Finally, twelve batches of PGAs from eight companies were identified and categorized into two types of strains (E. coli and Achromobacter sp. CCM 4824) using proteomic analysis. In total nine batches of five types of SSBAs (amoxicillin, cephalexin, cefprozil, cefdinir and cefaclor) from eight manufacturers were selected for investigation. Trace levels of PGA residual proteins ranging from 0.01 to 0.44 ppm were detected in six batches of different SSBAs which were far lower than the safety limit of 35 ppm reported by DSM, a manufacturer with expertise in the production of SSBAs by enzymatic processes. The developed FASP with LC-MS/MS method is superior to traditional protein assays in terms of selectivity, sensitivity and accuracy. Moreover, it could provide in-depth analysis of amino acid sequences and signature peptides contributing to assignment of the strain sources of PGAs. This method could become a promising and powerful tool to monitor enzymatic process robustness and reliability of this kind of SSBAs manufacturing.