RPLC/HILIC Orthogonal Impurity Profiling of Glucagon-like Peptide-1 (GLP-1) Analogs Using LC-UV/MS Workflows

Significance of the Topic

The rapid expansion of GLP-1 analog therapeutics increases demand for analytical strategies that deliver robust, high-confidence impurity characterization throughout development, manufacturing, and quality control. Traditional reversed-phase liquid chromatography (RPLC) is widely adopted for peptide analysis but can fail to separate impurities that differ only subtly in polarity or hydrophobicity. Integrating an orthogonal technique such as hydrophilic interaction liquid chromatography (HILIC) with LC-UV/MS workflows improves detection and resolution of such closely related impurities, strengthening method understanding and enabling reliable monitoring of critical quality attributes (CQAs) like deamidation.

Objectives and Study Overview

• Demonstrate a streamlined column screening workflow that leverages RPLC and HILIC orthogonality to enhance impurity profiling of GLP-1 analogs (exenatide model).
• Compare selectivity for clinically relevant impurities (proline insertion variants, deamidation products) across RPLC and two HILIC chemistries.
• Develop an MS-compatible HILIC method that can be integrated into existing LC-UV/MS workflows for routine impurity and degradant monitoring.

Methodology

Research-grade exenatide and synthetic deamidation standards were prepared at 1 mg/mL in pH 4.5 acetate buffer and diluted to 0.25 mg/mL for analysis. Forced degradation was performed in 0.1 M Tris-HCl (pH 8) at 37 °C to generate degradants. Separations used both RPLC and HILIC stationary phases with volatile, MS-compatible mobile phases based on ammonium formate and formic acid. Method optimization examined ionic strength, pH, and acid additives (formic, difluoroacetic, trifluoroacetic acids). Analytical criteria included chromatographic resolution, retention behavior, mass confirmation via LC-MS, and applicability for automated peak identification above impurity thresholds.

Used Instrumentation

• LC system: ACQUITY Premier System with Quaternary Solvent Manager (QSM)
• Detectors: TUV detector (λ = 280 nm) and ACQUITY QDa II Mass Detector (ESI+, full scan 450–1500 m/z, 5 Hz, capillary 0.8 kV, cone 15 V, probe temperature 600 °C)
• Chromatography software: Empower 3.8.1 for automated processing and reporting
• Columns and key conditions:
  • RPLC: ACQUITY Premier Peptide CSH C18, 130 Å, 1.7 µm, 2.1 × 100 mm, 60 °C, 0.1% formic acid aqueous/ACN mobile phases, flow 0.20 mL/min, 2 µL injections
  • HILIC (BEH HILIC): ACQUITY Premier BEH HILIC, 130 Å, 1.7 µm, 2.1 × 100 mm, 40 °C, ammonium formate-based mobile phases, flow 0.30 mL/min, 2 µL injections
  • HILIC (BEH Amide): ACQUITY Premier BEH Amide, 130 Å, 1.7 µm, 2.1 × 150 mm, 40 °C, 10 mM ammonium formate, flow 0.20 mL/min

Main Results and Discussion

• Raw material screening: RPLC-UV indicated ~99% purity for exenatide, but LC-MS revealed a co-eluting proline insertion impurity (+97.1 Da) representing ~40% of total ion intensity. This impurity was not resolved by RPLC despite extensive optimization (lower temperature, shallower gradients) without unacceptable method trade-offs.
• Orthogonal resolution with HILIC: BEH HILIC achieved baseline separation of the proline insertion variant (Rs = 2.94) using a relatively fast gradient, indicating HILIC’s ability to exploit hydrophilicity differences neglected by RPLC.
• Deamidation standards: Three synthetic standards were evaluated—exe-D (28D), exe-E (13E), and exe-DE (both sites). exe-DE (two acidic residues) separated readily from native peptide under all tested modes. The single-site variants exe-D and exe-E were challenging: RPLC required a very shallow, long gradient (0.5% ACN/min, 114 min) to reach partial resolution with peak fronting; BEH Amide HILIC co-eluted the two variants even with a 150 mm column; BEH HILIC provided partial resolution with a steeper gradient (1% ACN/min) in ~45 minutes, doubling throughput relative to the RPLC long-gradient run.
• HILIC method tuning: Increasing ionic strength reduced retention (salt masking effect); lowering pH reduced retention by changing protonation and electrostatic interactions. Acid additives markedly affected retention and selectivity: formic acid gave the narrowest peaks, least tailing, and best selectivity among FA, DFA, and TFA. The elution order of deamidation variants depended on additive—highlighting tunable selectivity in HILIC driven by multiple interaction mechanisms.
• Degradant profiling (forced degradation): Three major degradants were identified by LC-MS using the optimized BEH HILIC method: exe-D, a succinimide intermediate (-17 Da), and a species consistent with isoaspartic acid (+1 Da) that eluted separately from exe-E and showed ~3.9× the area of the aspartic acid variant, in agreement with known succinimide-mediated deamidation pathways.
• Workflow integration: Use of MS-compatible HILIC mobile phases and Empower CDS enabled automated peak identification by retention time and mass, flagging impurities above reporting thresholds (0.05% area) and supporting traceable reporting for QC and regulatory documentation.

Benefits and Practical Applications

• Orthogonal RPLC/HILIC profiling increases analytical confidence by revealing impurities that may co-elute in conventional RPLC-UV workflows.
• HILIC can significantly shorten run times for certain separations (e.g., deamidation variants) compared with ultra-shallow RPLC gradients, improving laboratory throughput.
• MS-compatible HILIC mobile phases allow direct integration with LC-MS detectors for structural confirmation and automated data processing, facilitating method transfer across development and QC laboratories.
• The approach supports impurity monitoring, forced-degradation studies, and routine release testing by combining complementary selectivity with mass confirmation.

Future Trends and Potential Applications

• Broader adoption of orthogonal chromatographic screening (RPLC + HILIC) early in method development to reveal hidden impurities and guide selectivity choices.
• Increased automation and integration with data systems and informatics for accelerated method scouting, automated peak annotation, and QC reporting.
• Combination with higher-resolution MS, ion mobility, or multi-dimensional LC (2D-LC) to further dissect complex peptide variants and low-abundance degradants.
• Continued evolution of HILIC stationary phases and mobile-phase chemistries to fine-tune selectivity for specific CQAs (e.g., deamidation, isomerization, acylation).
• Application of machine learning to predict orthogonal column performance and optimize method conditions, reducing experimental burden.

Conclusion

Integrating HILIC into LC-UV/MS impurity-profiling workflows provides orthogonal selectivity that reveals and resolves peptide impurities not readily separated by RPLC alone. Using MS-compatible mobile phases and modern CDS automation enables confident identification, monitoring, and reporting of clinically relevant impurities such as deamidation products and insertion variants. The combined RPLC/HILIC strategy improves method robustness, speeds certain analyses, and facilitates transferable workflows for development and quality laboratories.

References

1. Yu M.; et al. Battle of GLP-1 Delivery Technologies. Advanced Drug Delivery Reviews. 2018;130:113.
2. Han D.; Birdsall R.E.; Bhiwankar N. Accelerating Method Development and Manufacturing of GLP-1 Analogs with LC-UV/MS. Waters Corporation Application Note 720008800. May 2025.
3. Neale Q.; et al. Chromatographic Properties of Deamidated Peptides with Asn-Gly Sequences in Proteomic Bottom-up Experiments. Journal of Chromatography A. 2024;1738:465513.