Session: Parallel session 5 - New topics in OMICs

Evaluation of a timsTOF-Omnitrap prototype for the middle-down characterization of antibody repertoires

Simon OLLIVIER^1,2, Jean-François GREISCH⁴, Dimitris PAPANASTASIOU³, Albert J. R. HECK^1,2, Danique M. H. VAN RIJSWIJCK^1,2, Athanasios SMYRNAKIS³, Dina SCHUSTER^1,2, Jan FIALA^1,2, Florian BUSCH⁴, Mariangela KOSMOPOULOU³, Stuart PENGELLEY⁵, Oliver RAETHER⁵

¹Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and Utrecht Institute for Pharmaceutical Sciences, Utrecht University, Padualaan 8, 3584 CH Utrecht, Netherlands
²Netherlands Proteomics Center, Padualaan 8, 3584 CH Utrecht, Netherlands
³Fasmatech, Athens, Greece
⁴Bruker Switzerland AG, 8117 Fällanden, Zurich, Switzerland
⁵Bruker Daltonics GmbH & Co. KG, Fahrenheitstrasse 4, 28359 Bremen, Germany

Introduction

Antibodies are used to recognize and neutralize immunological threats in vivo, either endogenously by the immune system or as therapeutics. However, their sequence variability in certain regions, and high homology in others make their analysis challenging. With bottom-up proteomics the protein origin of peptides may easily be lost. Middle-down approaches retain this specificity—but require multi-stage (MSⁿ) and/or alternative fragmentation approaches, as typical collision-induced dissociation (CID) MS² is insufficient. We evaluate the potential of a novel timsTOF-Omnitrap prototype to characterize monoclonal antibodies (mAbs) and profile human serum antibodies.

Methodology

Antibodies were digested to release the fragments antigen-binding (Fabs) containing the variable regions of interest. The Fabs from three mAbs were then analyzed on a Bruker timsTOF Ultra mass spectrometer, modified to integrate an Omnitrap platform. For each, multiple charge states were submitted to electron-capture dissociation (ECD), electron-induced dissociation (EID), slow-heating CID and beam-type CID, as well as MSⁿ workflows. LC-MS compatibility was evaluated using a Bruker nanoElute 2.

Results

On all three Fabs and on the different charge states, electron-based MS/MS gave more sequence information compared to collision-based approaches. ECD fragmentation produced mostly c-type fragment ions, while EID produced both c- and a-type ions—giving complementary information. CID spectra were overall less informative, but interestingly yielded high-intensity sequence tag ions that we further analyzed by MSⁿ. We finally show that the timsTOF-Omnitrap prototype can be used in LC-MS to generate personalized antibody repertoires, alike to those described in recent studies by our group.

Conclusion

The timsTOF-Omnitrap affords advanced options for middle-down sequencing of antibodies and profiling of human antibody repertoires, soon to be merged for the de novo sequencing of serum antibodies.