Category: Formulation and Quality
Purpose: During manufacture of recombinant biotherapeutics, proteins from the host cells are secreted into the cell culture. Despite multiple purification steps, some of these host cell proteins (HCPs) can be present in the final product. These residual HCPs may have detrimental effects on the safety and efficacy of the drug product, therefore, the levels of any co-purifying HCPs must be determined prior to lot release.
The current regulatory gold-standard for assessing HCP levels in recombinant biotherapeutics is a sandwich immunoassay, or enzyme-linked immunosorbent assay (ELISA). This necessitates the generation of suitable anti-HCP antibody reagents, which often require considerable time and money to develop. Traditional ELISA approaches can also only provide limited information, in that they offer only bulk quantitation of HCPs that are immunologically reactive to the generated antibodies, with no possibility of identifying individual proteins.
Ultra-high-pressure liquid chromatography (UHPLC) separation combined with high resolution - accurate mass (HRAM) mass spectrometry (MS), however, can identify and quantify HCPs on a protein-protein basis, enabling individual HCP monitoring throughout the purification stages. The major challenge is to overcome the huge intrasample dynamic range to successfully detect low ppm levels of residual HCPs within the concentrated biotherapeutic drug product.
Methods: Dynamic range capabilities of the Orbitrap mass spectrometer were assessed by spiking a monoclonal antibody (mAb) with low levels of the Universal Proteomics Standard (USP1). Proteotypic peptides were generated using an automated purification system with magnetic resin immobilised trypsin, under denaturing (70°C) and non-denaturing (37°C) conditions. Any undigested protein was removed via precipitation at 95°C for 10 min. Peptides were separated by UHPLC using a Acclaim™ VANQUISH™ C18 UHPLC 250 x 2.1 mm (2.2 µm) column over a 90 min linear gradient. Data were acquired by both data-independent acquisition (DIA) and data-dependent acquisition (DDA) using Chromeleon™ CDS Software with an Orbitrap-based mass spectrometer. MS parameters were optimised to maximise HCP detection. Protein identification was performed using Proteome Discoverer 2.4 and Biognosys Spectronaut™ software for DDA and DIA analysis respectively.
Results: Due to the complex nature of recombinant monoclonal antibodies, non-denaturing digest conditions result in a largely intact biotherapeutic, but well proteolyzed residual HCPs. Removal of any non-digested protein with heat-treatment prior to MS analysis dramatically reduces the dynamic range of the sample, improving DDA MS analysis in terms of the number of detected proteins and number of unique peptides per HCP.
To maximize protein coverage and extend MS detection further into the sample dynamic range analysis was also performed using DIA. The DIA data analysis algorithms within Spectronaut, and its integrated search engine Pulsar, enables DIA data mining with and without a spectral library for identification and quantification of low abundance HCPs within a biotherapeutic drug product.
The magnetic resin immobilised trypsin kit simplified automated and reproducible protein digestion, for both denaturing and non-denaturing conditions, when used with the automated purification system.
To bring the HCPs into the detectable range of the mass spectrometer for DIA analysis, and thus enable detection of individual HCPs at the low ppm level in the final drug product, a column overloading concentration of biotherapeutic must be injected. The Acclaim VANQUISH C18 UHPLC column used in this study continued to deliver high efficiency and excellent separation even under these conditions, maximising protein coverage and enabling comprehensive data mining with HRAM-MS.
Conclusion: Orbitrap-based MS, combined with automated sample digestion, enables low ppm level HCP detection in a single-dimension UHPLC separation.