AUTOMATING SUBUNIT MAB ATTRIBUTE SCREENING FOR IN-PROCESS MONITORING OF AN AUTOMATED HIGH-THROUGHPUT MULTI-PARALLEL BIOREACTOR

Importance of the Topic

• Subunit-level characterization delivers critical quality attribute (CQA) insights into monoclonal antibody light chain, heavy chain and detailed glycoform profiles.
• High-throughput in-process monitoring ensures consistent product quality across varied bioreactor conditions and accelerates process development decisions.

Objectives and Study Overview

This study establishes a streamlined workflow combining automated sample preparation with rapid LC-MS subunit analysis to monitor mAb CQAs across multiple parallel bioreactors.

• Evaluate subunit attributes across five Ambr® 250 reactors with varied media, feed strategies and inoculation densities.
• Implement robotics to minimize manual steps and reduce processing time.
• Provide comprehensive analytics to inform process development decisions.

Instrumentation

• Ambr® 250 High Throughput Bioreactor System (Sartorius Stedim NA).
• Andrew+ Pipetting Robot and Extraction+ device (Waters Corporation).
• FabRICATOR® (IdeS) enzyme for subunit digestion (Genovis).
• BioAccord™ UPLC-MS with BioResolve™ RP mAb Polyphenyl column.
• waters_connect™ Intact Mass Application for data processing.

Methodology

• Sample Collection and Bioreactor Conditions
  • Five parallel Ambr® 250 reactors with two media types (A, B), two feed levels (1, 2) and low/high cell density.
  • Harvested on days 4, 6, 8, 10 and 12 for each condition.
• Automated Sample Preparation
  • Protein A purification in 96-well format using Andrew+ robotics.
  • FabRICATOR® digestion followed by DTT reduction to generate LC, Fd′ and Fc/2 subunits.
  • Total prep time for 48 samples under 3 hours.
• LC-MS Analysis
  • 4.5 min reversed-phase UPLC gradient at 60 °C.
  • Data acquired in positive ESI from m/z 400–7000.
  • Automated deconvolution and quantitation via waters_connect™ software.

Key Results and Discussion

• Workflow reproducibly generated high-quality subunit data across all bioreactors with minimal operator input.
• Major Fc glycoforms (fucosylated, afucosylated, Man5, C-terminal lysine) displayed distinct temporal trends linked to media and feed variations.
• Sialic acid species levels showed notable differences in specific reactors, indicating process impact on terminal glycosylation.
• Glycation on LC and Fd′ fragments remained consistently low (<0.1%), confirming robust culture conditions.

Benefits and Practical Applications

• Rapid, high-throughput CQA screening supports faster upstream process optimization.
• Automation reduces manual error and labor, improving consistency and throughput.
• Subunit-level insights enable targeted adjustments to media and feed strategies for desired product profiles.

Future Trends and Potential Applications

• Integration of real-time analytics and feedback control for adaptive bioprocessing.
• Extension of automated subunit workflows to other biologics and deeper post-translational modification profiling.
• Advances in miniaturization and multiplexing will further increase screening capacity.

Conclusion

The combined automation and rapid LC-MS subunit workflow demonstrated efficient, reproducible in-process monitoring of mAb CQAs across multiple bioreactors. This approach accelerates process development, ensures consistent quality and provides actionable insights for media and feed optimization.

References

• Koza S, Hanna C, Jiang Z, Yu YQ. Automated HT Analytical-Scale Monoclonal Antibody Purification Using Production-Scale Protein A Affinity Chromatography Resin. Waters Application Note 720007861; 2023.