Monitoring Intact Glycoprofiles and Spent Media Metabolites in Samples from Sartorius Ambr 250 High Throughput Bioreactor System to Support Upstream Process Development

Significance of the Topic

In biopharmaceutical upstream process development, rapid and integrated monitoring of critical quality attributes (CQAs) such as monoclonal antibody (mAb) glycosylation and cell culture performance parameters is essential for efficient optimization. Traditional workflows require separate assays for protein modifications and media metabolite profiling, leading to longer turnaround times and increased resource use. The implementation of a unified LC-MS platform with automated sample preparation can streamline data acquisition, improve consistency, and support data-driven decisions in real time.

Objectives and Overview

This study demonstrates a combined workflow for direct analysis of intact mAb glycoprofiles and spent media nutrient and metabolite patterns from CHO cell cultures run on the Sartorius Ambr 250 high-throughput bioreactor system. Key goals were to:

• Develop a single analytical platform for simultaneous glycan distribution and small-molecule profiling.
• Automate sample preparation for both intact protein and media analysis.
• Validate glycan quantitation directly from clarified media and compare with protein A-purified samples.
• Correlate MS or UV responses with conventional titer measurements.
• Generate multivariate data to support process optimization.

Methodology and Instrumentation

CHO cell cultures producing an IgG mAb were monitored in parallel bioreactors with varied media and feed conditions. Samples were collected on days 4, 6, 8, 10, and 12.

• Sample Preparation: Automated with Andrew+ Pipetting Robot and OneLab workflows. Clarified media were filtered and serially diluted for intact mass (1:20) and spent media (1:400) analyses. Protein A purification was performed on selected samples for comparison.
• LC-MS Conditions: BioAccord LC-MS System with reversed-phase separations using 0.1% formic acid in water (A) and 0.1% formic acid in 90% ACN/10% IPA (B). Intact protein gradients (200–4000 m/z) and metabolite acquisition range (50–800 m/z) were applied in positive and negative modes.
• Data Handling: glycoform deconvolution with MaxEnt1, titer trends via XIC and UV 280 nm in waters_connect interface, export back to Ambr software.

Main Results and Discussion

• Glycan Distribution: Five major glycoforms (G0F, G0F/G1F, G1F, G1F/G2F, G2F) were directly profiled from clarified media with clear TIC/XIC separation. Protein A-purified samples yielded comparable glycan percentages, supporting the use of direct analysis to save time and reagents.
• Titer Correlation: MS response (summed XIC of major charge states) and UV absorbance correlated strongly with Cedex Bio HT titer measurements (R2 = 0.954 and 0.902, respectively), with MS offering greater specificity at low concentrations.
• Metabolite Profiling: Approximately 100 nutrients and metabolites were detected and classified (amino acids, sugars, lipids, etc.). Trends such as tryptophan depletion and choline accumulation illustrated culture performance and feed efficiency.

Benefits and Practical Applications

This unified workflow delivers:

• A single compliant-ready platform for glycoprofiling, titer estimation, and spent media analysis.
• Automated, one-step sample prep reducing labor and variability.
• Rapid turnaround (<10 min per sample) suitable for high-throughput process monitoring.
• Seamless data integration with Ambr software for real-time decision support.

Used Instrumentation

• BioAccord LC-MS System
• Andrew+ Pipetting Robot with OneLab automation
• Sartorius Ambr 250 High-Throughput Bioreactor
• ACQUITY UPLC I-Class PLUS System and Tunable UV Detector
• waters_connect data interface

Future Trends and Applications

Looking ahead, this integrated approach can be extended by:

• Incorporating additional post-translational modifications and low-abundance variants.
• Implementing real-time analytics and feedback control in bioreactors (PAT integration).
• Applying machine learning to multivariate LC-MS data for predictive process modeling.
• Adapting workflows to other biologics platforms, including viral vector and cell therapy production.

Conclusion

The combination of automated sample prep and the BioAccord LC-MS platform enables high-quality, high-throughput monitoring of mAb glycosylation, titer, and spent media profiles. This streamlined workflow supports upstream process development by providing actionable insights faster and with reduced operational complexity.

References

1. H Shion et al. INTACT Mass™ – Rapid Mass Confirmation and Purity Assessment of Biotherapeutics. Waters Application Note 720007547, 2022.
2. YW Alelyunas et al. Monitoring Nutrients and Metabolites in Spent Cell Culture Media Using the BioAccord LC-MS. Waters Application Note 720007359, 2021.
3. SM Koza et al. Automated High-Throughput Analytical-Scale Monoclonal Antibody Purification. Waters Application Note 720007861, 2023.
4. YW Alelyunas et al. High Sensitivity Intact Monoclonal Antibody HRMS Quantification. Waters Application Note 720006222, 2018.