by Jeremy Cook, Ph.D, Marketing Manager, MOBILion Systems
Many protein-based therapeutics contain glycans as part of their structure, requiring thorough characterization and systematic batch control. However, as mass spectrometry instrumentation has become faster and more sensitive, the time and resolution of such analyses have become severely limited by the liquid chromatography (LC) methods used for separation. In this study, we demonstrate how these difficulties can be overcome by coupling hydrophilic interaction liquid chromatography (HILIC) with high-resolution ion mobility-mass spectrometry (HRIM-MS) and streamlined software, to comprehensively analyze N-glycans with substantially shorter HILIC run times, without compromising resolution.
Protein glycosylation is a critical quality attribute for biotherapeutics because of its direct impact on product safety and efficacy. A key difficulty is the high heterogeneity of these glycans (resulting from their non-template-driven biosynthesis), meaning that minor changes to the manufacturing process can result in significant changes to the glycosylation profile.
As a result, there is a need to separate and identify glycans with high resolution on a routine basis. This is typically achieved using LC separation, with detection and quantitation either with fluorescence-labeling or mass spectrometry. However, these methods have difficulty in resolving isomeric glycans, and the complete reliance on liquid-phase separation results in lengthy methods that restrict sample throughput.
Fortunately, the development of HRIM-MS is now allowing these challenges to be tackled. HRIM-MS uses moving electric fields on a printed circuit board to rapidly transport gas-phase ions along a long path in a lossless fashion, allowing them to be separated with very high resolution and in a timeframe of 2–5 minutes, without loss of sensitivity.1
In this study, we show how appending a fast HRIM-MS run to a HILIC analysis allows the HILIC run time to be shortened from 60 minutes to 10 minutes, while maintaining or even improving resolution for a set of N-glycans derived from Aranesp®, a biotherapeutic used to treat anemia resulting from chemotherapy treatment or chronic kidney disease. We complement this approach with software that allows feature-annotated heatmaps to be created for quicker and more simple data interpretation.
Experimental
Glycan samples were prepared from Aranesp® (Darbepoetin alfa, Amgen, Thousand Oaks, CA). N-Glycans were released using PNGase F (New England Biolabs, Ipswich, MA) and then labeled using a GlycoWorks RapiFluor-MS NGlycan Kit (Waters Corp., Milford, MA), following the manufacturer’s protocols. Samples were analyzed using a 1290 Infinity UHPLC system and AdvanceBio Glycan Mapping HILIC column (Agilent Technologies, Santa Clara, CA) with HILIC gradients of 10, 30, and 60 minutes (n = 4 replicate injections), followed by HRIM separation and detection using a MOBIE® commercial equivalent unit (MOBILion Systems, Chadds Ford, PA) coupled to a 6545XT QTOF (Agilent Technologies). Byos® software (Protein Metrics, Cupertino, CA) was used for data processing and report generation.
Results and Discussion
A total of 52 N-glycan compositions released from Aranesp® were analyzed by LC-HRIM-MS with HILIC gradients of 60, 30, and 10 minutes, followed by detection and relative quantitation. All three gradients show comparable relative glycan quantitation values and coverage, with the peak area RSDs being <5% for most assigned glycans, albeit higher for the 10-minute gradient. Encouragingly, even with the shortest HILIC run of 10 minutes, the separating effect of HRIM-MS was more than adequate to compensate for the reduced HILIC resolution. Figure 1 shows an example of this, where one HILIC peak for a single glycan composition is resolved into a cluster of several ion-mobility peaks, suggesting the separation of glycan isomers.
Figure 1: (A) Overlay of extracted ion chromatograms (XICs) for Aranesp® glycan 5_0_3_1_6 (m/z = 1113.09 +3) obtained using 60-, 30-, and 10-minute HILIC gradients. (B–D) Subsequent replicate overlays (n = 4) of the extracted ion mobiligrams (XIMs) for the three peaks shown in (A).
Even with the advantages of adopting LC-HRIM-MS for glycan analysis, analysis of such high-dimensional data has up to now been time-consuming. However, in this study we used software that resolves this issue, by providing a list of identified and assigned 4D features, along with an interactive heatmap (Figure 2) that can be used to show extracted ion chromatograms and easily inspect, adjust, and add feature assignments. It is anticipated that this should streamline LC-HRIM-MS data processing, aiding adoption within high-throughput workflows.
Figure 2: Example LC-HRIM-MS heatmap for the analysis of one of the N-glycan compositions, showing the unique arrival time profiles for each m/z value resulting from different glycan isomers and gas-phase conformations. Each dot represents an N-glycan signal, and the user interface allows data on the Byos HRIM Glycan workflow to be called-up by selecting a region of interest.
Conclusions
In this article, we have shown that adding HRIM-MS to traditional HILIC workflows allows run times for glycan analysis to be dramatically shortened without impacting performance. Importantly, the consistent HRIM-MS peak profiles obtained at increasingly short HILIC ramp times suggests that it can compensate for the reduced HILIC resolution, allowing rapid yet thorough characterization. This is aided by efficient analysis of the resulting 4D data, enabling rapid “fingerprinting” of glycosylation profiles for biotherapeutics.
We have also shown the potential for HRIM data to allow separation of glycans that co-elute in the HILIC run, suggesting that separation of glycan isomers on the basis of their ion-mobility arrival time is achievable. In the future, collision-cross section (CCS) values could be calculated and used to provide an extra layer of confidence for glycan assignments, and overall arrival time distributions could be used to establish the robustness of the bioprocess and ensure product consistency.
For more information about HRIM, please visit https://www.mobilionsystems.com/.
References
- L. Deng, Y.M. Ibrahim, A.M. Hamid, S.V.B. Garimella, I.K. Webb, X. Zheng, S.A. Prost, J.A. Sandoval, R.V. Norheim, G.A. Anderson, A.V. Tolmachev, E.S. Baker, and R.D. Smith, Ultra-high resolution ion mobility separations utilizing traveling waves in a 13 m serpentine path length structures for lossless ion manipulations module, Anal. Chem., 2016, 88: 8957–8964.