Mass-Based Fraction Collection (MBFC) for Scalable Purification in Pharma and Biopharma Applications

Mass-Based Fraction Collection (MBFC) for Scalable Purification in Pharma and Biopharma Applications — Summary

Significance of the topic

Mass-directed fraction collection integrates mass spectrometry (MS) with preparative high-performance liquid chromatography (prep HPLC) to substantially increase selectivity and confidence in isolating target compounds from complex matrices. In pharmaceutical and biopharmaceutical workflows, where samples contain many structurally related byproducts, relying on UV or refractive-index detection alone can lead to co-collection of impurities and lower yields of pure target material. Mass-based triggers allow precise discrimination by exact mass, adduct form, and charge state, enabling scalable and reproducible purification that is critical for compound characterization, lead optimization, and production-scale isolation.

Study aims and overview

This work demonstrates the capabilities of the Agilent InfinityLab Pro iQ Series Mass Detector integrated with Agilent preparative LC systems and OpenLab CDS software to perform mass-based fraction collection (MBFC). The objectives were to show how formula-based targeting, multi-trigger logic (including exclusion/NOT logic), and multi-charge-state monitoring improve collection selectivity and purity across representative small-molecule and biomolecule examples, and to highlight workflow flexibility for diverse purification scenarios.

Methodology

• System configuration: Experiments were performed on an Agilent 1290 Infinity II Preparative LC/MSD platform coupled to the Agilent InfinityLab Pro iQ Plus LC/MS detector and managed by Agilent OpenLab CDS software.
• Chromatography: Preparative reversed-phase separations were run on an Agilent Prep 100Å C18, 10 × 50 mm, 5 μm column. Typical preparative flow rates and delay coils were used to synchronize MS detection and fraction collection.
• MS targeting: Targets were defined by molecular formula or monoisotopic mass; expected adducts and charge states could be specified. OpenLab CDS auto-calculated m/z traces to generate summed SIM traces for a unified collection trigger.
• Triggering strategies: Up to four independent mass-based triggers were available, with configurable thresholds per trigger and boolean exclusion (NOT) logic to pause collection when specified impurity signals exceeded a threshold. Per-sequence override parameters enabled sample- or run-specific adjustments.
• Test cases: Examples included small-molecule dyes with diverse adduct behavior, caffeine with a coeluting impurity, and a peptide (angiotensin I) monitored across multiple charge states (+1 to +5).

Instrumentation used

• Agilent 1290 Infinity II Preparative Binary Pump (G7161B)
• Agilent 1290 Infinity II Preparative Open-Bed Sampler/Collector (G7158B)
• Agilent 1260 Infinity III Diode Array Detector WR (G7115A) with 0.3 mm preparative flow cell
• Agilent 1260 Infinity III Isocratic Pump (G7110B)
• Agilent 1290 Infinity II Preparative Column Compartment (G7163B)
• Agilent 1290 Infinity II MS Flow Modulator (G7170B)
• Agilent 1260 Infinity II Delay Coil Organizer with delay coils sized for 4–8 mL/min (G9324A)
• Agilent InfinityLab Pro iQ Plus LC/MS (G6170A)
• Column: Agilent Prep 100Å C18, 10 × 50 mm, 5 µm (part 446905-802)
• Software: Agilent OpenLab CDS v2.8 FR2 or later

Main results and discussion

• Formula- and adduct-based targeting: Defining targets by formula and expected adducts allowed automatic calculation of the relevant m/z values and consolidation into a single trigger. This approach avoids the need to choose a single adduct manually and increases robustness when ionization produces multiple species.
• Multi-trigger collection for heterogeneous responses: In a dye mixture where ion response varied by up to three orders of magnitude among components, three independent mass triggers with different thresholds enabled selective collection for each dye without collecting baseline noise. This maintained separation of components that would be difficult to isolate with a single fixed threshold.
• Monitoring multiple charge states: For peptides like angiotensin I, the software generated SIM traces across multiple charge states (+1 to +5) and summed them to form a unified trigger. Charge states that dominated the total signal (e.g., +2 and +3) drove collection, improving reliability without pre-selecting a single ion.
• Exclusion (NOT) logic to avoid contamination: Using a NOT trigger to exclude an impurity mass prevented fraction collection while the impurity signal exceeded its threshold. In the caffeine example, collection was delayed until the impurity fell below threshold, increasing measured purity of the collected fraction from ~93% to >99% on reanalysis.
• Operational flexibility: Per-run overrides for target mass, thresholds, and other parameters allowed adaptation within a sequence — useful for handling sample heterogeneity or scaling up multiple similar fractions with minor differences.

Benefits and practical applications

• Higher selectivity and purity: Mass-based triggers discriminate coeluting species that have similar UV responses, reducing rework and downstream purification steps.
• Improved yield confidence: Summed SIM triggers across adducts/charge states reduce missed collections caused by unexpected ionization behavior.
• Scalability across chemistries: The approach is applicable to small molecules, natural product extracts, synthetic reaction mixtures, peptides, and oligonucleotides where multiple charge states occur.
• Workflow efficiency: Built-in software features (multi-triggering, exclusion logic, per-sample overrides) reduce method development time and allow automated, robust fractionation during sequence runs.

Future trends and potential applications

• Tighter integration of MS intelligence with preparative automation — e.g., adaptive thresholds and real-time feedback for dynamic fraction windows based on peak shape and purity prediction.
• Expansion to high-throughput and continuous-flow purification workflows, where rapid decision-making by the MS can improve throughput and reduce solvent consumption.
• Application to biopharma modalities beyond peptides (e.g., oligonucleotides, conjugates) where multi-charge state monitoring and adduct handling are critical.
• Use of machine learning on combined UV/MS chromatographic data to predict optimal collection windows and impurity exclusion settings, reducing manual tuning.

Conclusion

Mass-based fraction collection using the Agilent InfinityLab Pro iQ Series Mass Detector and OpenLab CDS demonstrably increases selectivity, purity, and robustness in preparative HPLC workflows. Key strengths include formula-based targeting with adduct/charge-state handling, multi-trigger logic including NOT exclusion, and per-run override flexibility. These capabilities translate into higher-confidence isolations across small-molecule and biomolecule purifications and can streamline scale-up and repetitive purification campaigns.

References

1. Agilent Technologies. Harness the Power of Pro. Brochure, publication number 5994-8330EN, 2025.
2. Agilent Technologies. Improve the Productivity of Purification Workflows. Application note, publication number 5994-8532EN, 2025.