Automated Workflow for Peptide Mapping of Pembrolizumab on the Agilent 6545XT AdvanceBio LC/Q-TOF

Importance of the topic

Peptide mapping by LC/MS/MS is a cornerstone analytical technique for the in‑depth characterization of therapeutic monoclonal antibodies. It delivers sequence-level confirmation, site-specific identification of post‑translational modifications (PTMs), and disulfide‑bond mapping—information that is essential for ensuring product safety, efficacy, and lot‑to‑lot consistency. Automated, high‑quality peptide mapping workflows reduce variability, increase throughput, and provide the molecular detail required to monitor critical quality attributes (CQAs) during biologics development and manufacturing.

Aims and overview of the study

This application note demonstrates an end‑to‑end, automated peptide‑mapping workflow for pembrolizumab, integrating automated sample preparation, reversed‑phase LC separation, high‑resolution Q‑TOF mass spectrometry, and targeted bioinformatics. The goals were to: confirm the primary amino acid sequence including N‑ and C‑termini, verify the engineered S228P mutation, characterize low‑abundance PTMs (oxidation, deamidation, glycoforms, pyroGlu), and establish comprehensive disulfide linkage mapping in both reduced and nonreduced states.

Methodology

Sample preparation

- Pembrolizumab was supplied at 25 mg/mL and diluted to 6 mg/mL in 20 mM ammonium bicarbonate.
- Automated sample handling used the Agilent AssayMAP Bravo platform to perform denaturation, (reduced and nonreduced) treatments, heat denaturation, and tryptic digestion.
- Typical digestion conditions: denaturation/heat at 70 °C for 5 min, trypsin addition (15 ng/µL) and enzymatic digestion at 37 °C for 45 min. Reduced samples included DTT; nonreduced omits DTT.
- No manual desalting step was performed; early LC flow was diverted to waste to reduce buffer interference.

Chromatography and mass spectrometry

- Reversed‑phase LC on an Agilent Altura Peptide Plus column (2.1 × 150 mm, 2.7 µm) with 0.1% formic acid in water (A) and 0.1% formic acid in acetonitrile (B). Flow rate 0.4 mL/min, column 50 °C, injection 10 µL; gradient spanned low organic to ~80% B across the run (peptide signals concentrated ~30–75 min).
- Detection used an Agilent 6545XT AdvanceBio LC/Q‑TOF with Dual JetStream ESI. MS1 acquisition covered ~200–3,000 m/z; MS/MS CID fragmentation was applied with data‑dependent precursor selection and charge‑state preference targeting peptide isotopes (2+ > 3+ etc.). Tuned collision energies and isolation widths optimized peptide fragmentation and site localization of modifications.

Data processing

- Agilent MassHunter BioConfirm v12.1 performed peptide identification, PTM localization and disulfide mapping.
- Search parameters included peptide isotope model, MS1 ±10 ppm and MS2 ±20 ppm tolerances, CID fragmentation, up to two missed cleavages, peptide length 3–70 aa, and up to two variable modifications per peptide. For disulfide mapping, cysteine linkages were entered in sequence manager to generate and search theoretical disulfide‑linked peptides.

Instrumentation

- Agilent AssayMAP Bravo Protein Sample Prep (automated digestion and cleanup)
- Agilent 1290 Infinity III Bio LC system (high‑speed pump, multisampler with thermostat, multicolumn thermostat)
- Agilent Altura Peptide Plus column, 2.1 × 150 mm, 2.7 µm
- Agilent 6545XT AdvanceBio LC/Q‑TOF with Dual JetStream ESI source
- Agilent MassHunter data acquisition and Agilent MassHunter BioConfirm 12.1 for data analysis
- Agilent VWorks for workflow automation orchestration

Main results and discussion

- Sequence coverage: The workflow achieved 100% amino‑acid sequence coverage for pembrolizumab after trypsin digestion, with robust MS/MS confirmation across light and heavy chains, including unambiguous N‑ and C‑terminal peptide assignments.
- Engineered mutation (S228P): The peptide containing residue 228 in the Fc hinge was identified and MS/MS fragmentation confirmed the serine‑to‑proline substitution. The spectrum provided residue‑level confirmation of the intended modification that stabilizes IgG4 behavior.
- Oxidation: Methionine oxidation at HC position 252 (and other methionine residues) was detected and site‑localized by characteristic +16 Da mass shifts in MS/MS fragments (example: y‑ion shifts observed between native and oxidized forms). Extracted chromatograms showed baseline separation of oxidized and native peptides; relative quantification of oxidized species was performed from ECCs.
- Disulfide mapping: Nonreduced and reduced analyses allowed detection of intact disulfide‑linked peptides and reduced peptides, respectively. All 16 expected disulfide linkages (12 intrachain and 4 interchain) were detected and assigned using BioConfirm, with distinct chromatographic features for disulfide‑linked species.
- Workflow robustness: Automated sample handling eliminated manual digestion variability and yielded high‑quality MS/MS spectra enabling confident detection of low‑abundance species without manual desalting, aided by diverting early LC flow to waste to minimize ion suppression.

Benefits and practical applications

- The integrated automated approach increases reproducibility, throughput, and data quality for peptide mapping, which is crucial for release testing, CQA monitoring, biosimilarity assessments, and stability studies.
- High sequence coverage and confident PTM localization support regulatory documentation (e.g., ICH Q6B requirements for terminal and sequence confirmation).
- Detection and quantitation of low‑abundance modifications (oxidation, glycoforms) and complete disulfide mapping enable informed process control and product comparability evaluations.

Future trends and potential applications

- Further automation and standardized sample prep will be coupled with higher‑throughput LC/MS platforms to meet growing sample volumes in process characterization and QC labs.
- Advanced acquisition strategies (e.g., PRM/DIA hybrid methods) and alternative fragmentation chemistries may improve sensitivity for labile PTMs and isomeric modifications.
- Machine‑learning‑assisted spectral interpretation and automated reporting will accelerate decision making and reduce manual review burden.
- Integration with intact and middle‑level MS workflows can provide complementary structural context (e.g., glycoform distribution, intact mass) for thorough biotherapeutic characterization.

Conclusion

The demonstrated workflow combines automated AssayMAP sample preparation, robust reversed‑phase LC, and high‑resolution Q‑TOF MS with dedicated bioinformatics to deliver comprehensive peptide mapping of pembrolizumab. Outcomes include full sequence confirmation, terminal peptide verification, site‑specific PTM identification and quantitation (including low‑abundance oxidation), S228P mutation confirmation, and complete disulfide linkage assignment. The approach supports routine CQA monitoring and provides a scalable platform for development and QC laboratories.

Reference

1. Li X, et al. High Throughput Peptide Mapping Method for Analysis of Site‑Specific Monoclonal Antibody Oxidation. J Chromatogr A. 2016;1460:51–60.
2. Agilent Technologies. Automation for LC/MS Sample Preparation: High Throughput In‑Solution Digestion and Peptide Cleanup Enabled by the Agilent AssayMAP Bravo Platform. Application note 5991‑2957EN.
3. Wong DL, Chen J. An Integrated Workflow for Peptide Mapping of Monoclonal Antibodies. Agilent Technologies application note 5991‑7811EN; 2017.
4. National Center for Advancing Translational Sciences. Pembrolizumab amino acid sequence. (GSRS/Ginas dataset accessed 2026).
5. Alhazmi HA, Albratty M. Analytical Techniques for the Characterization and Quantification of Monoclonal Antibodies. Pharmaceuticals. 2023;16(2):291.
6. BenchChem. A Deep Dive into Pembrolizumab: A Structural and Functional Comparison. 2026.
7. Patnaik A, et al. Phase I Study of Pembrolizumab (MK‑3475; Anti–PD‑1 Monoclonal Antibody) in Patients with Advanced Solid Tumors. Clin Cancer Res. 2015;21(19):4286–4293.