Mass spectrometry of crosslinked peptides

Proteins are central functional components of life, yet studying their structure and interactions remains a major challenge. The work I present in this thesis aims at addressing this challenge, advancing the identification of crosslinked peptides by analysing their mass spectrometric fragmentation behaviour and leveraging the knowledge gained for the development of advanced acquisition methods. First, I investigate the influence of the employed fragmentation method and parameters on peptides crosslinked with the most commonly used crosslinkers BS3 (bissulfosuccinimidyl suberate) and DSS (disuccinimidyl suberate). We show that the choice of fragmentation parameters greatly impacts the number of reliably identifiable crosslinks. The most significant factors impacting the crosslink identifications are fragment sequence coverage and acquisition speed. We developed a data-dependent decision tree acquisition method, choosing the best-performing parameters depending on analyte properties. We further established higher-energy C-trap dissociation (HCD) as the best overall single method for the analysis of BS3 / DSS crosslinked peptides. Second, I present a web-based tool that enables the visualisation, sharing, and analysis of mass spectrometric fragment spectra. While the spectra identification process is nowadays dominated by database matching algorithms, visualisation for manual verification, hypothesis testing and (re-)analysis of identified spectra remain important. Our community-standard-compliant tool xiSPEC renders high-quality vector graphics of annotated spectra in the browser, allowing the readily testing of hypotheses and sharing of proteomics data. It is available as a stand-alone online service and has also been successfully integrated into other tools. Third, I studied the fragmentation behaviour of linear peptides using ultraviolet photodissociation (UVPD), fragmenting analytes using a 213 nm UV laser. We analysed a wide variety of peptides using a specifically designed acquisition workflow and could show that the fragmentation behaviour of most linear peptides is relatively poor but improves with higher aromatic amino acid content. This work laid the foundation for the later development of a UVPD-cleavable crosslinker (UCCL). Fourth, crosslinking MS was further advanced by introducing a cleavable moiety into the crosslinker reagent. This allowed for advanced acquisition regimes and showed promising results, especially for proteome-wide crosslinking. We analysed the fragmentation behaviour of the most commonly used CID-cleavable crosslinker disuccinimidyl disulfoxide (DSSO) showing that the major improvements result from improved sequence coverage by better fragmentation of the crosslinked peptides and less so from knowledge of the individual peptide masses. We further identified a lack of specificity and sensitivity in MS3-based approaches as a major hindrance to unlocking the full potential of MS-cleavable crosslinkers. Fifth, the combination of insights gained from our UVPD fragmentation study with our work on CID-cleavable crosslinkers lead to the development of a UVPD-cleavable crosslinker. The mixed fragmentation of peptides and crosslinker inherent to CID-cleavable crosslinkers hampers the targeting of the individual peptides for separate fragmentation. Employing UVPD at 213 nm as an orthogonal fragmentation technique to specifically target a chromophore in the crosslinker largely increased the sensitivity and specificity of selecting individual peptides for MS3. The general concept of orthogonal cleavage between the linker and analyte presented could also improve the analysis of other linked analytes. In summary, high-quality fragment spectra are crucial for the challenging analysis of crosslinked peptides, as they are the heart of the crosslink identification process. Advanced integrated workflows comprising functional crosslinkers, intelligent MS acquisition methods, and data analysis algorithms will continue to play a pivotal role in further enhancing crosslinking MS technology.

Author dc:creator

• Kolbowski, Lars Bernhard

Advisor dc:contributor.advisor

• Rappsilber, Juri

Licence dc:rights.uri

• https://creativecommons.org/licenses/by/4.0/

Language dc:language.iso

en

Identifier URI

https://doi.org/10.14279/depositonce-18283

OAI identifier oai:identifier

oai:depositonce.tu-berlin.de:11303/19485