An improved EasyPep sample preparation method for enrichment and quantitation of host cell proteins

Importance of the topic

Host cell proteins (HCPs) are critical impurities in biotherapeutic production that can affect drug safety and efficacy. Analytical methods must overcome the large dynamic range between abundant therapeutic proteins and low-level HCP contaminants. The improved EasyPep sample preparation workflow featuring protein A depletion and semi-tryptic elution offers a promising strategy to enrich and accurately quantify HCPs by LC-MS/MS.

Objectives and study overview

The primary goals of this study were to compare four distinct sample preparation workflows for HCP analysis in trastuzumab drug substance and to develop a targeted MS assay for 28 high-risk Chinese hamster ovary (CHO) HCPs. The workflows included native digest with and without protein A depletion, and EasyPep-based digestion with and without protein A depletion plus trypsin elution. Quantitative performance, HCP coverage, and dynamic range reduction were evaluated.

Methodology and instrumentation

The four workflows started from 1 mg of affinity-purified trastuzumab. The EasyPep approach utilized magnetic protein A beads for antibody removal followed by semi-tryptic digestion to recover bound HCPs. Native digestion controls were prepared with and without pre-depletion. Heavy-labeled AQUA peptide standards (65 peptides, 200 fmol) were spiked into 0.5 µg of digest. Chromatography was performed on a Thermo Scientific Dionex UltiMate 3000 RSLCnano system coupled to a Q Exactive Plus for data-dependent acquisition and parallel reaction monitoring (PRM). A proof-of-concept targeted MSn assay was implemented on the Thermo Scientific Stellar MS platform using a 30-minute active gradient and unscheduled/scheduled MS2 and MS3 acquisition methods. Data processing and assay building were conducted in Skyline with Prosit-predicted spectral libraries and PRM Conductor for transition refinement.

Results and discussion

The protein A-depleted EasyPep workflow identified twice as many HCPs (16 unique proteins) compared to native digest (3 unique proteins) and outperformed the other two methods in overall HCP coverage. Quantification of 65 AQUA peptides by PRM revealed superior accuracy and dynamic range reduction in the protein A-depleted EasyPep samples. Dilution curves demonstrated reliable detection down to atto-mole levels, and LOD/LOQ distributions confirmed lower limits of quantitation for both MS2 and MS3 modes. The Stellar MS proof-of-concept assay accurately detected unknown HCPs in null-antibody control samples, validating its applicability for broad dynamic range analyses.

Benefits and practical applications

• Enhanced HCP identification and quantitation accuracy compared to conventional workflows.
• Improved dynamic range management by depleting high-abundance monoclonal antibody.
• Robust targeted assay for critical CHO HCP biomarkers enabling process monitoring.
• Compatibility with existing LC-MS/MS platforms and software tools for streamlined implementation in QA/QC environments.

Future trends and opportunities

• Integration of in silico spectral prediction and AI-driven data analysis to accelerate assay development.
• Automation of sample preparation workflows for high-throughput HCP testing.
• Expansion to multi-attribute methods combining HCP profiling with other product quality attributes.
• Real-time process analytical technologies (PAT) incorporating continuous LC-MS monitoring of HCPs.

Conclusion

The modified EasyPep sample preparation method with protein A depletion and semi-tryptic elution significantly enhances HCP enrichment and quantitation performance. Coupled with heavy-labeled peptide standards and targeted MSn acquisition on advanced platforms, this workflow provides a powerful tool for comprehensive HCP monitoring in biopharmaceutical development and manufacturing.

References

1. Jones et al. (2021). “High-risk host cell proteins (HCPs): A multi-company collaborative study.” Biotechnology and Bioengineering, 118(8):2870–2885.
2. Pino et al. (2020). “Matrix-matched calibration curves for assessing analytical figures of merit in quantitative proteomics.” Analytical Chemistry, 19(3):1147–1153.
3. Heil et al. (2021). “Comparison of unit resolution versus high-resolution accurate mass for parallel reaction monitoring.” Journal of Proteome Research, 20(9):4435–4442.