Back to Top

Ask about this


How to Determine Glycan Profiles of Biopharmaceuticals from Peptide Mapping Data

Stuart Pengelley, Bruker Daltonics GmbH & Co. KG

Eckhard Belau, Bruker Daltonics GmbH & Co. KG

Waltraud Evers, Bruker Daltonics GmbH & Co. KG

Anja Resemann, Bruker Daltonics GmbH & Co. KG

Detlev Suckau, Bruker Daltonics GmbH & Co. KG

Jul 27, 2021


Glycosylation is a common critical quality attribute (CQA) of therapeutic proteins and needs to be characterized during product development. Typically, this analysis is conducted after enzymatic release and tagging of the glycans followed by fluorescence and MS detection. However, that approach loses information about the glycan position which is important for complex biologics such as fusion proteins; and it requires dedicated method setups and experience far beyond the classical peptide mapping analysis.

Here we describe a workflow to identify glycan compositions directly from tryptic peptide maps [1] acquired with the timsTOF Pro and PASEF® in combination with analytical LC and the new VIP-HESI ion source. It employs a glycan search method from peptide mapping data and uses previously identified aglycons as mass tags. Using this setup, we identified 36 different glycan compositions from NISTmAb tryptic digest which correlate well with those reported in the literature [2]. The results described here were generated directly from peptide mapping data without the need for a dedicated glycan laboratory workflow, thereby retaining the information about which peptide is glycosylated.


Characterization of therapeutic proteins such as monoclonal antibodies requires a broad range of analyses including the full confirmation of the protein sequence as well as the detection and identification of post-translational modifications such as protein glycosylation. A common workflow to confirm the entire protein sequence is the peptide mapping approach, which combines pro-teolytic digests with RP-LC-MS/MS analysis in which, e.g., tryptic peptides are separated by reversed phase chromatography and further analyzed using high-resolution tandem mass spectrometry. N-linked glycans are subsequently analyzed in a second experiment comprising enzymatic glycan release, chemical labeling, LC separation on graphitized carbon or HILIC columns and MS analysis [2].

Glycopeptide analysis is typically applied in glycoproteomic experiments to identify the protein as well as the glycan using a specific fragmentation involving a collision energy stepping method [1]. The resulting glycopeptide MS/MS spectra provide both glycan and peptide fragments and allow to assign the peptide sequence as well as the corresponding glycan structure in one spectrum. Glycopeptides from therapeutical antibodies like NISTmAb, however, are derived from a tryptic peptide of known sequence and therefore can be assigned with reasonable confidence based on accurate mass alone.

When applied to glycopeptides, standard PASEF conditions used for peptide mapping preferentially cleave the glycosidic bond between the carbohydrate units, and subsequently peptide fragments are not observed. Such spectra are dominated by y-type glycan fragments attached to the intact peptide and lower molecular weight b-type glycan fragments (without peptide moiety – i.e., the aglycon - attached). Similar fragmentation patterns are observed in CID spectra from released glycans carrying a fluorescence label like RapiFluor at the reducing end (Figure 1). In the approach reported here, the aglycon of a glycopeptide is treated in a glycan database search like a mass tag at the reducing end. The peptide tag serves the same purpose for the ionization of the glycan as fluorescent labels like RapiFluor in the analysis of labelled glycans – the peptide has an even stronger ionization propensity providing a high analysis sensitivity.

The timsTOF Pro with PASEF [3] further increases the capability to acquire high quality MS/MS spectra from glycopeptides with high dynamic range despite the almost non-existent reverse phase separation of the glycopeptides with identical aglycon. In this work we evaluated the identification characteristics of the glycopeptide approach on the timsTOF Pro linked to the VIP-HESI ion source (Vacuum Insulated Probe – Heated ESI) – a heated ion source optimized for applications requiring high sensitivity.

Reagents and Equipment

Mass spectrometer:  timsTOF Pro mass spectrometer via a VIP-HESI ion source 


Sample Preparation

NIST Monoclonal Antibody Material 8671 (Merck) was reduced using DTT and TFE (25%) for 150 min at 56 °C and alkylated using IAA. Trypsin (Promega) digestion was performed overnight.

LC-MS Data Acquisition

Twenty μg tryptic digest were separated in a 60 min gradient on an Acquity CSH C18 2.1 x 100 mm 1.7 μm column (Waters) using an Elute UHPLC (Bruker). The UHPLC was interfaced with a timsTOF Pro mass spectrometer via a VIP-HESI ion source and peptides were analyzed by PASEF using the standard proteomics acquisition method, adapted by lowering the precursor intensity threshold to 400 counts.

Data Analysis

The raw data were initially processed using BioPharma Compass® 2021b (Bruker) with the Peptide Mapping method Tutorial NIST_mAb with small modifications resulting in classical peptide maps with comprehensive sequence coverage (not shown). Typically, the Fc-glycopeptides are eluting in a narrow retention time range early in the gradient. To reduce the number of MS/MS spectra submitted to the glycan search and computing time, the Rt range for data computing was limited to the glycopeptide elution range (5-15 min, Figure 4).

In a second processing step, all MS/MS spectra in the Rt range were submitted to a glycan search using the GlycoQuest search engine included within BioPharma Compass.

The Fragmentation Type CID byi4Cl was initially defined in “Admin Preferences”: Protocols/ProteinScape/Glycomics/GlycoFragmentationType. Only b-, y- and internal ions are selected. Maximum cleavages are set to 4, max crosslinks to 0.

The tutorial glycan search method N-glycan QTOF CID RapiFluor was adapted subsequently to the analysis of the previously known/established tryptic NISTmAb Fc glycopeptide EEQYNSTYR using the following parameters (Figure 2):

  1. Use GlycO as database as it is a rather condensed database containing all relevant glycans expected on a humanized IgG1
  2. Define the Reducing end: Reducing end mass: 1170.494166 (peptide mass of EEQYNSTYR -H2O was calculated using the Sequence Editor in BioPharma Compass) Reducing end name “fc” for Fc tryptic peptide
  3. Select Fragmentation Type CID byi4Cl
  4. Thresholds for result compilation were stringent to reduce the glycan list for confident search results: Score > 40, Fragmentation coverage [%] > 40, Intensity coverage [%] > 40


The glycan search of the MS/MS spectra yielded 36 specific glycan compositions of different glycan classes like complex, hybrid, and high mannose structures (Figure 3).

The glycopeptides were detected at 2 different Rt ranges: glycopeptides with neutral glycans and those with acidic glycan units (N-glycolyl-neuraminic acid) eluted after approx. 8 and 10 min, respectively (Figure 4).

The most intense glycan structure (G0F, h3n4f1) was detected with an absolute intensity of 161x103, lower abundant compositions were identified with intensities below 2x103 (Figure 5). These were mainly non-fucosylated neutral or acidic glycan structures or doubly fucosylated glycans like G1F2 or G2F2 (h4n4f2 or h5n4f2, Figure 6).

Compared to the results of a traditional analysis using released labeled glycans and HILIC separation [2], 27 out of 30 different compositions were found using our glycopeptide analysis method. The three missing compositions were h7n2, n7n3 and h8n5f1. The compositions h4n2, h4n3, h4n4f2, h5n3, h5n3g1, h5n4f2, h6n4g1 and h6n4f2 were additionally detected using the method described here (Figure 6). Differentiation of isomeric structures has not been further investigated.

The co-elution of glycopeptides always presents a challenge for LC-MS analysis, which typically results in much lower assignment rates of glycan compositions compared to labelled glycan analysis. Here, however, PASEF contributed the required high acquisition speed (> 100 Hz MS/MS) and sensitivity (ion mobility time and space focusing), whilst sensitivity and signal intensities were given a further boost by the VIP-HESI ion source, which is heated to 400 °C. The combined effect was that 36 glycan compositions were identified from a single aglycon using standard, well established peptide mapping methods, which correlates well with data published by an expert group using a dedicated released glycan approach [2].


  • Thirty-six different glycan compositions were identified from a NISTmAb tryptic digest by dedicated glycoanalysis of a selected peptide using a conventional peptide mapping dataset and the GlycoQuest search engine in the BioPharma Compass software.
  • The result compares favorably with previous studies using the established labelled-glycan approach [2]. However, a thorough isomer assessment still relies on MS/MS analysis and spectra library-based identification of labelled glycans.
  • The PASEF technology was key to acquire high quality MS/MS spectra from coeluting glycopeptides covering a dynamic range of 100:1. The setup included analytical RP-LC separation and the new VIP-HESI ion source. This new approach enabled glycosylation analysis for non-experts.
  • For more complex glycoproteins with multiple glycosylation sites, such as the SARS-CoV-2 S-glycoprotein [4], the approach can be repeated peptide-wise to obtain site-specific glycan composition profiles with great sensitivity and specificity.


[1] Hinneburg H, Stavenhagen K, Schweiger-Hufnagel U, Pengelley S, Jabs W, Seeberger PH, Varón Silva D, Wuhrer M, Kolarich D (2016). The art of destruction: Optimizing collision energies in Quadrupole – Time of Flight (QTOF) instruments for glycopeptide based glycoproteomics. J Am Soc Mass Spectrom, DOI: 10.1007/s13361-015-1308-6

[2] Hilliard M, Alley WR Jr, McManus CA, Yu YQ, Hallinan S, Gebler J, Rudd PM (2017). Glycan characterization of the NIST RM monoclonal antibody using a total analytical solution: From sample preparation to data analysis. mAbs, DOI: 10.1080/19420862.2017.1377381

[3] Meier F, Brunner AD, Koch S, Koch H, Lubeck M, Krause M, Goedecke N, Decker J, Kosinski T, Park MA, Bache N, Hoerning O, Cox J, Räther O, Mann M (2018). Online Parallel Accumulation–Serial Fragmentation (PASEF) with a Novel Trapped Ion Mobility Mass Spectrometer. Mol Cell Proteomics 17(12):2534-2545, 

[4] Gstöttner C, Zhang T, Resemann A, Ruben S, Pengelley S, Suckau D, Welsink T, Wuhrer M, Domínguez-Vega E (2021). Structural and Functional Characterization of SARS-CoV-2 RBD Domains Produced in Mammalian Cells. Anal Chem 93(17):6839-6847. doi: 10.1021/acs.analchem.1c00893. Epub 2021 Apr 19.


Be the first to comment on this protocol.

Submit a Comment

First Name *

Last Name *

Email *

Comments *

Premier Sponsor