Evaluation of peptide fractionation and native digestion as two novel sample preparation workflows to improve HCP characterization by LC-MS/MS.

Host cell proteins (HCPs) are the predominant class of impurities during manufacturing of therapeutic proteins. Previous reports have successfully shown that HCP characterization by LC-MS/MS ultimately leads to drug products of superior safety and quality. Here, we present two sample preparation strategies to approach the wide dynamic range required and compared them systematically to a standard protocol. First, we describe PreOmics fractionation as an effective 2D offline strategy. Second, we evaluate an alternative digestion approach specifically designed for purified antibodies - native (nondenaturing) digestion. Both protocols increased detection sensitivity as shown by two low level HCP models. Out of a 5 ppm spike of eight common HCPs into antibody product, all spiked proteins were positively identified. Additionally, by Universa Proteomics Standard 1 (UPS-1) spiking we obtained a comprehensive coverage of 77% below 10 ppm for the native digestion. Furthermore, we were able to detect 27% to 173% more HCPs in protein A elution pools of five different antibodies and to reject new concerns of HCP coprecipitation by pellet digestion. Although it encounters new challenges, the native digestion is very attractive due its simplicity and comparability to 2D workflows. However, for complex samples such as mock transfected cell culture fermentation, best results were obtained with peptide fractionation. This study highlights the advantages of both methods and their value to facilitate LC-MS/MS approaches to become an even more powerful tool for HCP profiling.